Smarter Phones: Journey to the Palm-Sized Computer

In the early 90s, a handy software developer took a stack of wood, and carved it into small blocks of various sizes. He carefully compared the weight of each block. And when he found one that felt pocket-sized, he taped a printout of a tiny monitor onto it. Then he tucked the block in his shirt pocket, and walked around with it, to see how it would feel to be attached to a device. He was imagining, in a not so distant future, where we'd all be doing the same thing. If you think that guy's name was Steve Jobs, you're wrong. His name was Jeff Hawkins, and he co-created the PalmPilot. When the iPhone hit the market in 2007, critics and competitors questioned whether a smartphone would succeed.

A decade later, the question is, how can a person succeed without one? Smartphones are ubiquitous. Their apps allow us to do pretty much anything. And the hardware running them says a lot about who we are. But as central as the iPhone has been to the rise of our mobile lives, it wasn't the catalyst. This is the epic story of how an earlier handheld device paved the way for the smartphone. And it's the story of a devoted team that stuck with that device for its entire journey. I'm Saron Yitbarek, and this is Command Line Heroes, an original podcast from Red Hat.

The smartphone concept has been around since Star Trek’s tricorder. In real life though, the concept first translated into cell phones in 1984, bulky things that looked like bricks during the 90s they got a bit smaller, small enough for Zack Morris to carry on Saved by the Bell—but they were still just used for phone calls. Remember phone calls? Nothing “smart” was happening on mobile phones, but there was another piece of technology gaining traction. It was called a PDA, a personal digital assistant, a mobile electronic device that acted as your personal information manager. We'll get to that in a moment, but at the time, the tech industry was way more focused on the personal computer, which we learned about in episode 3 when we looked at the Altair 8800.

Everyone was so caught up in what a personal computer was, this huge, big beige box sitting under your desk. They couldn't imagine that you'd carry this thing around in your pocket.

Ed Colligan was VP of marketing at a nascent mobile software company called Palm in the early 90s. Palm was founded by Jeff Hawkins, the guy who walked around with a block of wood in his pocket.

It was a big vision. It was the future of computing, of personal computing as handheld computing and that there would be more transactions done on handheld computers in the future than on desktop computers.

That's Donna Dubinsky, Palm’s CEO at the time.

I know today when I say that it sounds like, “Whatever, that's logical.” But believe me, it was not logical at the time.

We didn't understand why other people didn't understand it, because you know, where had computing gone, right? It had gone from computers that filled a room to mainframe computers to minicomputers, which were kind of misnamed to personal computers to desktop computers. And we saw the inevitable march of Moore's Law and more and more power and smaller, smaller packages.

Palm started out developing information management software for a PDA that Casio was making, called the Zoomer. They also made some synchronization software for Hewlett Packard's devices, but those first-gen PDAs weren't taking off. And then the whole personal digital assistant dream looked like a lost cause after the high-profile failure of Apple's efforts. The Newton—they were all too big, too heavy, and the software was too slow. But the Palm team wondered whether a new approach could change the game.

The original deal we worked on was with an operating system company, GeoWorks, and a hardware provider, Casio. However, what happened along the way was we figured out that industry architecture, if you will, that stacked up didn't make sense for handheld computing. What we learned was the right way to build these was with highly integrated hardware and software products. So, what changed our mind was born out of failure, essentially.

Palm figured that if they could build their own hardware, something that ran their software, they might be able to build a winning PDA.

No, no personal digital assistant actually wasn't our term.

Whoops, sorry Ed.

We didn't like PDA. We literally positioned it as a connected organizer. We effectively tried to change the way people looked at it, from being a whole new device that you need to adapt to, to being an accessory to your PC.

When Ed says connected, he means connecting the device to the computer, not connected to the wireless internet. That wasn't happening yet, but even being able to sync your Palm device with your personal computer was a step above what the competition offered.

They didn't connect with anything, so if you had an address book in your Sion and your Casio Wizard, it had nothing to do with your address book on your computer. Instead of a standard handheld standalone computer, we positioned it as an appendage of your computer, a window onto your computer, a way to take a little piece of your computer with you, which put the synchronization function at the center of the positioning.

In addition to building a mobile organizer device, Palm wanted to pair it with a synchronization device that you could plug into your computer to update information.

Which seemed very intimidating at the time. There's 25 people, hardware is very... It's capital intensive, you need a lot of investments, you need time, you need procurement, and it was quite an intimidating challenge, frankly.

That's Rob Haitani, Palm’s product manager. And this was Donna Dubinsky’s challenge to Rob’s team.

And we wanted to create from scratch an operating system, application software, hardware synchronization software, any one of these was effectively a company. So, for a little company to take on all of that at one time, and have it all work together, was really audacious as I look back.

This is why people go to Silicon Valley. It's like, okay, you're going to create an entirely new industry. The people like IBM and Apple and Microsoft have failed at this. So, what makes you 25 people think you can succeed? And you know nothing inspires people in Silicon Valley more than telling them they can't do something. We believed in Jeff's vision, and Jeff had a very clear vision. And, what I really... what really resonated to me, was very customer focused. He said, "Don't build technology, build a solution for a customer."

It was that customer-first attitude that inspired Jeff Hawkins to carve himself that gadget-sized block of wood.

So, he’d carry this thing around in his pocket, and they would walk into meetings, and he had this little wooden stylus, and he would pretend that he was trying to simulate what the experience would be like. So, we'd pretend to be writing on it, and people would give him these funny looks cause he's, you know... you're writing on a piece of wood, Jeff. I think that really gave him some insights into what it would really... the experience would feel like.

That little block of wood inspired some core design criteria. First, it had to fit into a shirt pocket. Second, it had to have a price tag of less than $300. Third, it had to be faster than pen and paper. You could instantly turn it on and use it. Here's Ed and Rob comparing Palm's approach to design versus the competition’s.

Everybody else was trying to create this standalone device and so they said, “Oh, we need an expansion card for more memory,” or they'd say, “We have to have a keyboard,” and we basically said, “No, we're going to hook it to the PC, and we're going to synchronize the information between the PC and this device instantly.”

Jeff was a guy who took a completely contrarian point of view. He said, the problem you're trying to solve is it's too slow. Then the conventional wisdom is to add a faster processor. He took the other approach saying, well, how do you make it faster for the customer? It's not only about the hardware, it's about if you make the software slim and light, then that will succeed and then that will have its own spiraling effect of: that will be small, it'll be lighter, it will have longer battery life. So we developed this philosophy of stripping down the number of steps and being super critical about efficiency.

They even gave their philosophy a name: “the Zen of Palm.” They wrote a whole design manifesto and everything. Designing the Palm OS platform involved new concepts and new ways of thinking that were different from designing for bigger machines. With PCs, the more features the better. But handhelds are a different creature. According to the Zen of Palm, handhelds should be all about the user. Features should be tightly focused and the device should be usable anywhere.

How do you fit a mountain into a teacup? You want to find the diamond in the mountain and put that in the teacup. The purpose of a design for a small screen is not to take a desktop PC, full-functional design, and cram it and miniaturize it. It was to take the nuggets that you really needed at any time and put that on the screen. And the way we addressed that was, we literally would step through every task and say what absolutely has to be on the screen.

Reducing the amount of buttons on any given page allowed for a smaller screen, and Rob's team drastically reduced the number of taps it took to execute function.

He literally would go through every screen and every function and see how he could reduce the taps. An example I like to use is that rather than three-tap—turning on a device, hitting a calendar app, hitting the date for today—this was one of the most common things you wanted to do. What if you could just press one button, and this button would turn on the device, take you to the calendar app, and show you today? It was a “today button.” And now, again, these sorts of things seem obvious, but at the time, this was quite radical. That was not how you interacted with devices. So, that took tremendous coordination between the hardware and the software.

Within 18 months, Palm had done the seemingly impossible. They had a prototype with beautifully synchronized hardware and software. And they'd done it with only $3 million. But there was one problem. It was their last $3 million. Palm had an amazing new handheld computing device, and they were flat broke. The Palm team's hail-Mary solution to keep on financing their new product was to sell the company to US Robotics in 1995. It was the only way they could get this new connected organizer, which they dubbed the PalmPilot into the hands of customers. And they knew there'd be customers. Here's CEO Donna Dubinsky, remembering what happened when Jeff Hawkins unveiled the “PalmPilot” at a tech conference.

We had Jeff up there on the stage, showing the device, but the moment that was the most powerful, and that got us a near standing ovation was, he brought out the cradle, and he put the device in the cradle, and he pressed the button. And you could see on the screen it synchronizing with the PC. And that blew people away. And they just spontaneously applauded. So it was very exciting to see how they got the core value proposition of what we were trying to sell.

Palm's goal at that launch was to convey to people how simple the device was to use. But they didn't have a big budget to produce their reveal. So, they got creative.

We came up with this crazy idea to have our mothers come, and help us launch it. So we had my mom, Jeff's mom, and Ed's mom. They had little hats we made up that said “Moms for Pilot.” They had little pins that said “My daughter's Donna,” “My son is Ed,” or whatever. And we had them take orders for Pilots on the spot. I bought my Pilot from Jeff's mom, or from Ed's mom, or whatever. To this day, people stop me and say, "I bought my Pilot from your mom." It was really, really fun. They had a blast, and it made for a memorable launch.

What the moms were signing orders for was a huge step forward in handheld computing. But keep in mind, this is happening in 1996. Ed Colligan and Rob Haitani again.

The product ran on two AAA batteries, for a month, ok. It had 128K of memory. You've probably never heard a K of memory. It had a display screen that was a black and white, you know, display.

We had a screen that was 160x160 pixels, which is microscopic. I mean, I've designed icons almost that size. So we had a very low-powered processor, and a very small amount of memory. So it had a 16 megahertz processor, 128K of RAM, and we had to make an operating system work under those constraints. Low-powered screen. It was not color, it was not even grayscale. It was a monochrome, 160-pixel screen, and then below it, we had a digitizer that was not a screen, but it was a digitizer, so you could write on it.

Palm added its handwriting recognition software, called Graffiti to the PalmPilot. Remember, that it didn't have a keyboard. If you wanted to write something on it, you used a stylus.

You wrote on a rectangle at the bottom, and you wrote letters on top of each other, and you had to learn a simplified alphabet.

The technology was new and smooth, but it wasn't without its problems. For example...

We had been shipping for a little while, and we suddenly started getting catastrophic failures in the field. Devices that went off, and just couldn't go back on, and we started getting these into our service department, and had to try to figure out what was going wrong.

This is why startups don't typically make hardware. It's very difficult, and we had this problem where people were losing data, and we couldn't figure out what was happening.

The team combed through all their documentation, back through their many change agreements and orders. They tried tracing it back to something that changed with their process. Still, they couldn't figure it out. Out of frustration, Donna got everyone together.

I put all the senior people in one room and almost locked the door and said, "You're not leaving here until you figure out what's wrong with this thing. Why is this happening?"

Eventually, the team realized there'd been a tiny change inside the machine, but not the kind you'd expect. It had nothing to do with the hardware components at all.

You know how when you take a battery cover off, or you put in batteries on a device, so the inside cover there, somebody added a sticker with some kind of a warning or something.

The source of this giant headache was a little warning sticker that had been added to the underside of the battery cover.

That sticker caused friction with the batteries, and they could get depressed in a way that disconnected the power. And there had been a software patch that had been loaded on, and that patch was lost when the power was disconnected.

The hardware team swapped out connectors for springs to bolster the batteries, a super-simple fix for a catastrophic failure.

The good news was it was all synchronized, so it was all backed up. And you realized that for people, it's a real light-bulb moment. The data is the value, not the device.

Maybe you're listening to this podcast on a smartphone right now. Take a look at it. Your phone is light-years ahead of those old PalmPilots. And yet, the basics of what you're using were all there in the Palm. Chris Dunphy was Palm's Director of Competitive Analysis.

It was this kind of amazing golden age. Palm launched to the market in 1996, with the PalmPilot, and it was the buzz everywhere. It was the cool thing to have, this little thing in your pocket that was a portable brain, and Palm was smart enough that they'd put out a developer SDK as kind of, almost, a side effect, and that took off. All these little niche markets had really cool little apps popping up for them. From everything from doctors to knitters. And people were in love with their devices, in love with their apps.

There was an existing community of developers making similar apps for Mac's desktops, and they hopped over to build an app ecosystem for Palm.

A lot of the original Palm developers weren't large companies. They were just small hobbyists who were doing projects in their spare time. They had some personal passion project, and their minds exploded when they started thinking, what is a computer, and a screen that you carry with you all the time? And it becomes an extension of your mind. And so many people had so many great ideas of how to take advantage of that software development kit, and write really, really cool things. And, it was really groundbreaking.

I mean, I know a lot of people think Apple invented the App Store, and the idea of apps on handheld, but actually the very first PalmPilot had a very early app store. It was a third-party app store, and very early developers who came in and did all sorts of creative applications that again, people could just sync onto their device.

The plan was to sell 100,000 units in the first year. And for the first six months, sales were stable. About 10,000 units a month. But then, things started to skyrocket.

And in fact by 18 months, we sold a million units, which was the fastest new product growth in American history at the time. I mean it was stunning growth. A million units in 18 months.

Palm had created an entirely new category of hardware. And the spoils were, theoretically, theirs for the taking. But then, unexpectedly, Palm's parent company, US Robotics was sold to another company, called 3Com. And the head of 3Com, who was influenced by the Microsoft business model, decided to license the Palm OS to companies that wanted to create PDAs of their own.

And obviously, it was a strategy that has succeeded very well from Microsoft in personal computers. We didn't think that was the right strategy for handheld computers. We felt they needed to be highly integrated devices, but they consistently felt that that was the wrong decision, and that what we should do is license the OS to all commerce. And we disagreed with that.

They believed in their vision. And so, right after they helped their parent company revolutionize the market, Jeff Hawkins, Donna Dubinsky, Ed Colligan, Rob Haitani, and others left Palm to form a new company. They called it Handspring. There, they would license the software they'd created, Palm OS, and load it onto their own handhelds. They had built a giant. And now, they were going to try to take it down, David-and-Goliath style, using its own OS. In 1999, the newly formed Handspring, free from those old parent companies released its own hardware, the Visor line of PDAs. And they ran on Palm OS. Ed Colligan remembers its public reception.

And sure enough, we took like 25% of the market, almost overnight.

Palm's hardware was hit by that move. So ironically, 3Com spun it off into the independent company the team had hoped for. In the meantime, the Palm OS was running on 90% of all handheld computers, not too shabby. In fact, for a short period, Palm was worth more than Ford and General Motors combined. People thought it'd become the next Microsoft. Meanwhile, Handspring had its own plans.

By the time we started Handspring, we started realizing that these devices ultimately would be communications devices and we built it with a hardware slot and the hardware slot was specifically with the idea to be able to experiment and integrate any kind of communication things—put in a pager card, put in a voice card, put in whatever, and that we would learn from that, and learn how to integrate communications and what would be important in the space.

We saw the smartphone coming, we saw that these things were all going wireless and we decided we wanted to figure out how to create that integrated device of both the PDA and the phone.

So Handspring got to work creating a smartphone. In the process, they replaced the stylus with the keyboard and named their new creation the Treo. While all this was in progress Jeff, Donna, and Ed met with another tech entrepreneur doing interesting things in the space. A guy by the name of Steve Jobs.

At that meeting, Steve got up on the board and drew out a Macintosh and he had all these things like photos and video and other things as satellite things off of the Macintosh and he said, "Our strategy is the Macintosh is going to be the center of everything and all these things are going to pivot around it." And that was iTunes, iPhoto, whatever. Right? And Jeff said, "Nope, that's not how it's going to work. How it's going to work is there's going to be a handheld computer and all these things are going to pivot off of it."

We know how this all turned out. Jeff Hawkins’ vision was actually closer to the truth, but at the time, early 2000s, Jobs was skeptical. The whole industry was skeptical.

I used to go into Sprint and Verizon, and these guys, and try to convince them that smartphones were going to be something. I know it's hard to believe today, but literally we'd sit down in meetings and they're like, eh, these new-fangled devices or you know ... who's going to do email on something in your pocket. And I'd go, "Well, I really think it's going to happen."

But while they waited for the world to catch up, they had another more pressing problem. It threatened the future of Handspring, Palm, and just about everything. In 2001, the tech bubble burst, stocks plummeted, money was suddenly scarce and investments dried up. So in another hail-Mary, this time to manufacture the Treo, Handspring merged back into Palm. I know all the back and forth is making me a little dizzy, too. The Treo became Palm's powerhouse product and the most popular smartphone on the market. But of course by that point, the Palm OS had started to show its age. New players had entered the market. Companies like RIM with its Blackberry.

Wireless was becoming a thing and experts seriously doubted whether the Palm operating system was a good fit for the next generation of devices. So in 2005, Palm shipped its first Treo without the Palm OS. They built that Treo around Windows mobile. By 2007, Palm had become a hardware company with no operating system of its own. The future that the Palm team wanted to build seemed to be rolling on without them.

Palm needed help and they got it in the form of John Rubinstein, the man who developed the iPod at Apple, just as Apple released the iPhone in 2007, Rubinstein came on board at Palm as their new head of product development. Two years later, the Palm team had a new device, the Palm Pre and a new OS called Web OS. They launched at CES in 2009. Some called it the best tech keynote ever. Here's Ed Colligan onstage at the event.

And it's called the Palm Web OS and we're very, very excited to bring it to you today. It was built with developers in mind. The whole thing is built on industry-standard web tools. If you know HTML, CSS, and JavaScript, you can develop applications for this platform.

No one had ever seen anything like Web OS, it laid the groundwork for the whole smartphone experience we take for granted today. In fact, iOS and Android gleaned a lot from its features. Features like multiple synchronized calendars, unified social media and contact management, curved displays, wireless charging, integrated text and web messaging, unintrusive notifications. You could upgrade it easily just by putting it into dev mode and you could receive over-the-air updates. Web OS was an amazing achievement that no other company could match. Unfortunately, it wasn't enough.

I think we did a phenomenal job with that, but it was just too little, too late, because at that point Apple had launched the iPhone. They executed really, really well and so all power to them, but I think they were hugely influenced by what we had done and to this day, I mean until, like, OS X or whatever on the iPhone, that was the first time they'd actually caught up with all the features that were in the Web OS.

But Ed thinks the real killer was another phone.

The killer blow was Google and Android and their ability to not have to make money off of it, other than search.

Google basically gave Android away for free. It was a problem for Microsoft's Windows phone and for the Palm pre/Web OS combo.

And we did not have that business model, and it just hugely undermined us and, and there was really no way to recover from that.

After creating a whole new tech category with the PalmPilot, dominating mobile software with Palm OS, building the first smartphone, the Treo, reinventing mobile OS with Web OS, after all those innovation and iterations, Palm was sold to HP in 2010 and then later to LG. In 2012, HP released open Web OS built on top of a Linux kernel.

Once it was open source Web OS became the underlying OS for tons of other smart devices, TVs, watches, and the Internet of Things. And that old debate over fusing hardware and software, well, I'll let Donna Dubinsky settle things.

They're virtually indistinguishable from each other. You can't have great hardware and terrible software and you can't have great software and terrible hardware. The question is almost nonsensical. They have to be together. You know, you carry these things on you all the time. It's a highly integrated device. People don't even know where the hardware ends and the software begins and that's as it should be.

In Jeff Hawkins’ case the hardware and the software began with that small block of wood tucked away in his shirt pocket. That simple block of the right shape and size has launched a fleet of millions, perhaps billions, of smartphones 25 years later.

Command Line Heroes is an original podcast from Red Hat. Go to our website for some amazing bonus material we dug up on Palm and Web OS. Redhat.com/commandlineheroes. And hey, while you're there, sign up for our newsletter. I'm Saron Yitbarek. Until next time. Keep on coding.