The Story Behind Apple's NewtonS

When the first whispers of an Apple tablet started going around, we'd found it amusing that Apple hired back an old Newton PDA developer. Now that the mythical device is closer, we've been reminiscing more about Apple's original tablet.

In the late 1980s, Apple appeared to be in the middle of a resurgence. John Sculley had forced out the volatile Steve Jobs in 1985, and a cadre of older, more experienced executives focused on building the Apple and Macintosh brands. The company was beginning to grow complacent, working to protect Macintosh revenues at the cost of interoperability and new technology.

John Sculley, Apple's CEO, had toyed with the idea of creating a Macintosh-killer in 1986. He commissioned two high budget video mockups of a product he called Knowledge Navigator. Knowledge Navigator was going to be a tablet the size of an opened magazine, and it would have very sophisticated artificial intelligence. The machine would anticipate your needs and act on them.

One video showed a college professor working with the device to effortlessly prepare a lecture while the computer created the graphics and simulated different models. Sculley believed that such a device would be the next big thing in the computer industry, and he desperately wanted Apple to be the company to develop it.

Handwriting Recognition

Steve Sakoman was especially cognizant of Apple's dependence on the Macintosh. While Sakoman was at HP, he worked with alternative input devices centered around different configurations of keypads. He hoped that eventually keyboards would be rendered obsolete and people would use touchscreens to interact with computers equipped with handwriting recognition software.

HP was uninterested in handwriting recognition, and Sakoman was assigned to help design the HP Portable, one of the earliest laptops.

Steve Jobs hired Sakoman in 1984 to help work on a laptop version of the Macintosh after the successful release of the HP Portable. When Jobs left Apple, these laptop plans were scrapped, and Sakoman helped lead the teams creating the Mac Plus, Mac SE, and Mac II.

He found the work uninteresting, however. He wanted to leave Apple to work on handheld computers, and he recruited Jean Louis Gassée to lead a brand new company that would be bankrolled by Lotus founder, Mitch Kapor. The plan fell through, since it appeared that Apple would probably sue the nascent company.

The Newton Project

To keep the talented Sakoman from defecting, Gassée proposed creating a skunk works project to create an Apple handheld computer. Gassée got permission to start the project from Sculley (without telling him what was being researched), and Sakoman set to work.

The first thing he did was select a name for the project. Because Apple's original logo had a rendering of Isaac Newton sitting beneath an Apple tree, Sakoman decided to name the project Newton.

Sakoman gathered a team of engineers (including Finder coauthor, Steve Capps) and moved into an abandoned warehouse on Bubb Road in Cupertino, in the same vein of the Macintosh team almost a decade before.

The Newton researchers started work on a specification for a tablet computer.

Sakoman's end goal for Newton was to create a tablet computer priced about the same as a desktop computer. It would be the size of a folded A4 sheet of paper and would have cursive handwriting recognition and a special user interface.

To run the enormously demanding handwriting recognition software, the tablet would have three AT&T Hobbit processors. Sakoman and Capps feared that the project would balloon in scope and ultimately create something so expensive that it would flop.

The worst of Sakoman's fears came true. The engineers had no restrictions on size or cost, so they started piling features onto the product. The first Newton would be an A4 sized slate with a hard drive, an active matrix LCD, and infrared for high speed, long distance networking. Named Figaro, the product would cost well over $6,000 and wouldn't be released until 1992.

Debugging Software

For two years, the Newton researchers had toiled away, creating mockups and sample software. Several prototype tablets had been assembled and were running very buggy software.

The major hanging point was handwriting recognition. It was incredibly difficult to create handwriting recognition software that was able to adapt to different writing styles. Fate intervened in a bizarre way: One night while Apple VP of board relations, Al Eisenstat, was in Moscow, he heard frantic knocking on his door. When he answered, he saw a nervous programmer scanning the hallway to see if he was being followed. The programmer handed Eisenstat a floppy disk containing handwriting recognition software, then quickly left.

When Eisenstat returned to the States, he gave the code to Gassée, who in turn gave the code to the Newton team. The recognition was remarkably accurate, and it actually adapted to learn different letter shapes, so it learned how to read each user's handwriting.

Financial Problems

By the end of 1989, there was a sense of foreboding around Apple as growth slowed markedly. Between 1987 and 1989, Apple's sales had grown by over $2 billion, but in 1990, sales barely grew (by just $10 million). Apple's strategy of releasing high-end products with incredibly profit margins (30% higher than most other PC manufacturers) was beginning to fail, and Sculley knew it.

Gassée had long defended the practice, as most of the profits were pushed into his division. He actually took steps to prevent Apple from losing the high profit margins by entering the low end. Claris had started a project, called Drama, to create another brand to sell low-end Macs. Drama got as far as contracting with Nissan Design to create the enclosures before Gassée killed the project. He reasoned that consumers would be willing to pay much more for the Macintosh experience.

After the release of the enormously successful Macintosh Classic (Apple's cheapest Mac until the iMac hit US$999 in 2000) and the meteoric growth of Compaq, Gateway, and Dell, Gassée's strategy started developing cracks. It was long assumed that Gassée would succeed Sculley as CEO; instead he became the scapegoat for Apple's stunted growth.

When Michael Spindler was named COO (years after the enormously popular Del Yocam was forced out by Gassée), Gassée felt snubbed, and on March 2, 1990, he resigned. Gassée took Sakoman with him to found Be Inc., where he would help design the BeBox.

More Trouble for Newton

After Sakoman left, Newton was in trouble. In the Advanced Technologies Group (a think tank inside Apple), a former Gassée fief, a product similar to Figaro was being developed. Marc Porat had created a number of documents describing a concept called "agents", small programs that would travel around a network gathering information, then present it to the user or even act on it without intervention. Apple would provide the software (called Paradigm, for the paradigm shift it would predicate) and hardware (Pocket Crystal), while outside companies would provide the wireless networks and content.

Pocket Crystal appeared to pose a significant threat to the Newton project. Two Mac heavyweights were working on the project, Andy Hertzfeld and Bill Atkinson. Sculley sent Larry Tesler to assess the Newton project and see if there was anything that could be used in Pocket Crystal. Tesler had been a researcher at Xerox's Palo Alto Research Center (PARC), but he left shortly after Jobs toured the labs. At Apple, he helped design the Lisa interface and was named an Apple Fellow.

Tesler was impressed with the Newton project. By that time, Figaro had three processors, a battery that ran nonstop for weeks at a time, and it weighed eight pounds. He was a little uneasy with the $8,000 price - as much as a Unix workstation. The Newton engineers responded to his incredulity by making the case that there could be no compromises if Figaro was to succeed.

Capps had created a mock up of the Newton software using HyperCard and showed it to Tesler, who was immediately impressed. He threw his support behind the Newton and quickly took control of the group. To insure Newton's continued survival, Sakoman had the team create a more detailed demonstration to show Sculley and the rest of the board, along with a marketing plan from Michael Tchao, the marketing manager who had joined the Newton project weeks before Sakoman left, containing specifications and prices of possible Newton products.

Sculley was enamored with Newton, especially Newton Intelligence, which allowed the software to anticipate the behavior of the user and act on those assumptions. For example, Newton would filter an AppleLink email, hyperlink all of the names to the address book, search the email for dates and times, and ask the user if it should schedule an event.

Pocket Crystal fell out of favor with Sculley and was quickly spun out as General Magic. The company enjoyed some early success, until it languished and eventually sold most of its intellectual property to Microsoft in 1998.

From Research Project to Marketable Product

Sculley set a ship date for Newton, ending its status as a research project. A Newton handheld that cost less than $1,500 would have to be ready by April 2, 1992.

Tchao came up with three Newton models. The first one to be released was a larger version of the Figaro tablet, renamed Senior. Senior would by 9" x 12" and cost around $5,000. A year after the Senior debuted, two smaller Newtons would be released. A midsize model, which was quickly killed, would measure 6" x 9" and cost less than $2,000. The smallest model was Junior, which would be 4.5" x 7" and cost around $500.

Most of the engineering and marketing staff favored Junior, but Tesler pushed Senior. He believed that an underpowered device would set a negative precedent for future Newton products. But Tchao believed that the expensive Senior would relegate Newton to early adopters and vertical markets, not the consumers that Apple had the most sway over.

The two factions were at each other's throats. Tesler favored Senior, since it included a lot more advanced features. His favorite was Senior's ability to connect to five other units simultaneously via infrared.

To the Junior engineer's great amusement (and derision), the feature would not work in a room with fluorescent lighting. Most offices would have to turn off their overhead lights to network their Newtons!

Tchao believed that the infrared was a prime example of how Senior was wrong for Newton. It was focused on bringing new technologies to consumers, not creating a viable product.

The conflict simmered for months until Tchao found an ally in Sculley. The two were sharing a ride on Mike Markkula's private jet. Tchao launched into an impassioned speech making the case for Junior and how Senior would actually harm Newton. Sculley concurred and instructed Tesler to shelve Senior and work to release Junior.

Focus on 'Junior'

Engineers pushed ahead with the software. They worked with the same fervor the Mac team had under Steve Jobs. Not only would Junior change Apple, it would change the world. Beyond that, Apple engineers were often the recipients of huge bonuses if they stayed ahead of schedule (the engineers working on Star Trek were given a free vacation to a resort in Mexico for successfully demonstrating their project). It was not unusual for the engineers to be working between fifteen and twenty hours every day.

Operating system and language development was farmed out to another group in ATG. Based out of Cambridge and led by Ike Nassi, the team worked on a language inspired by the ease of use of the Smalltalk language and the power of C++. Named Ralph, for author Ralph Ellison, the language would be licensed to other software developers and would be used as the primary language for Pink, Apple's planned operating system.

Good-bye Hobbit, Hello ARM

Development began to bog down as it became clear that the language would not run efficiently on the Hobbit processor that Sakoman had selected three years before.

The Hobbit was being developed by AT&T as a low power RISC processor meant to be used in switching relays and embedded applications. The early specifications that Sakoman had created used three Hobbit processors. To cut costs and improve battery life, that had been reduced to one, but the processor was not powerful enough to run the demanding software.

Besides that, AT&T asked that Apple invest $1 million in the completion of the product. Tesler backed out of the deal and went to a small British company named ARM along with Ford, Hobbit's other major customer.

ARM had been started by Acorn, the computer manufacturer. Acorn had made a number of phenomenally popular personal computers in Britain during the early 80s. It reached the apex of its success when it released the BBC Microcomputer, the official computer for the popular program, The Computer Programme. The machine was based on the 6502 processor, which was fine for home applications, but it was reaching its limits with the advent of computer graphics.

Acorn had hoped to move to the new Intel 80286, but Intel refused to give Acorn the samples necessary to design a new computer. Outraged, Acorn started a project led by Roger Wilson to create a RISC processor to power its new line of Archimedes computers running RISC OS.

In 1985, the team had completed its first machine. The ARM (Acorn RISC Machine) CPU had performance comparable to the Motorola 68000 used in early Macs, but with half as many transistors, which meant it consumed less power and generated less heat.

Apple "discovered" ARM, and on September 8, 1990, the company bought a 43% stake in ARM, which was split off from Acorn and subsequently renamed Advanced RISC Machines. Apple would use the ARM6 chip in the Newton. Tesler was named to ARM's board of directors as Apple's representative. (ARM is now the most popular processor in the world, with 1.5 billion chips sold a year.)

The new processor was not only faster than Hobbit, it was also more efficient. An ARM equipped Junior could be much smaller, since it required less cooling space.

Unfortunately, it wasn't fast enough to run the taxing Ralph language. The language was stripped down to run on the ARM and renamed Dylan. Apple quickly ran into trademark trouble because of the new name, which officially stood for Dynamic Language. Bob Dylan threatened to sue Apple for using his name without permission. Apple and Dylan eventually settled out of court for an undisclosed amount.

A Broad Base

Apple didn't want to take the Newton plunge alone and started looking for a partner. Ideally there would be several different Newton devices available at launch from different manufacturers. Beyond that, Apple had little experience building consumer devices, so it would be much less expensive to contract manufacturing to an outside company.

After contacting Matsushita and Sony, Apple settled on Sharp. Sharp already had a presence in the organizer market with its Wizard, so the company knew how to sell such devices. Besides that, Sharp was the largest LCD manufacturer in the world. Sharp would build Apple's Junior (which was named MessagePad for release) and its own version, called ExpertPad.

Apple also reached agreements with Motorola and Siemens to produce their own versions. (shortly after Motorola released the Newton-powered Marco, the company would jump ship to General Magic and its Magic Cap operating system.)

John Sculley, Evangelist

Sculley was losing interest in day to day operations at Apple. By 1991, he was only actively involved in Newton and Pink. That year Sculley campaigned hard for Bill Clinton (Sculley had been a lifelong Republican until then) and sat next to Hillary Clinton during his first State of the Union speech.

Because of his campaigning, Sculley became the de facto spokesman for the computer industry (like Steve Jobs in 1981), and he granted frequent interviews.

Sculley decided to espouse the value of the Newton during his keynote speech at the CES in Las Vegas on January 7, 1992. Sculley never mentioned the Newton by name, but he made the case that "personal digital assistants" (PDAs) would one day become commonplace.

The larger theme of the speech was digital convergence, and he predicted that the market for devices and content would reach $3 trillion by the end of the decade.

Delays

Sculley was painfully shy (he often ate lunch alone table because he disliked making small talk), but he was an excellent public speaker. The press ate up his words, and he planned to announce the Newton during CES Chicago that summer along with a demonstration of early versions of the MessagePad, which was not even close to being ready for the April 2 deadline Sculley had set.

This was the first of several ship date changes for MessagePad.

Steve Capps began a massive campaign to identify and squash bugs so the MessagePad could be demonstrated at CES Chicago. The engineers put in longer days and worked on fixing thousands of bugs. The software was quickly declared to be in alpha status, which meant that no new features could be added to the software; it would only be fixed and optimized.

Despite their efforts, MessagePad was still very unreliable. The team was forced to drop Ralph altogether because of the slow progress being made on the language. Lots of the software had to be totally rewritten. Tesler practically lived in the office, spending 18 hour days planted in front of his computer hacking away on code.

Preparations for the CES demonstration in Chicago were underway, and the engineers dove even deeper into their work. Capps, who owned a large home outside Cupertino, had a high speed ISDN connection installed at his home so he could work remotely. He was so productive that two other Newton engineers joined him there.

There were an incredible number of bugs to fix. Oftentimes units became too hot to handle because of faulty power managers. The software was not very reliable either. As the team practiced demonstrations of key features like faxing and beaming information via infrared, the MessagePad failed more often than it succeeded.

The Demo

Luckily the demonstration went well.

Apple rented space to show off the MessagePad in Chicago. All of the MessagePads were tethered to Macs - they were too unreliable to run independently. The engineers demonstrated some shape recognition and showed off the user interface, which was based on a notebook paradigm, not the desktop of the Mac's Finder.

All of the demonstrations went very well with no hitches. But Newton was still not in the clear.

The result was a flood of positive press for the MessagePad. The entire computer industry was rushing to bring similar products to the market. General Magic gave its first public demonstrations of Magic Cap weeks after the Newton introduction, and companies such as Microsoft and Amstrad announced that they would release similar products.

Apple's Consumer Electronics Division

Sculley had expected to stay at Apple for only five years, but Apple was growing so rapidly during 1989 that Sculley stayed on (though his wife soon returned to Connecticut).

Michael Spindler had become President and COO and called most of the shots. He engineered a major reorganization shortly after he was promoted. Sculley had created two major divisions when he wrested control from Jobs in 1985, research and development and marketing. Spindler broke Apple into more than half a dozen smaller pieces that focused on specific markets. Newton was rolled into the Personal Interactive Electronics (PIE) division that contained all of Apple's consumer electronics.

The division was headed by former Phillips executive Gaston Bastiaens. His crowning achievement at Phillips was to release the CD Interactive (CD-I) console. CD-I was essentially a video game console that used CD-ROMs. The package cost well over $1,000 and never became popular. Eventually, the standard was relegated to interactive kiosks in European department stores.

CD-I demonstrated Bastiaens' major flaw - he was willing to release pricey products with poorly defined markets. PIE would soon release a line of digital cameras, scanners, speakers, and even a portable CD player. Most of the products were priced well above the competition, and they all failed to gain a foothold.

Bastiaens was enthusiastic about Newton. He gave the team a new deadline to finish the MessagePad - July 29, 1993. Bastiaens made the decision not to release the date to the press, which was quickly labeling the MessagePad as vaporware.

Newton Untethered

Another CES was coming up in Las Vegas on January 8, 1993, and the team was working as hard as ever to get ready. This time the MessagePads would not be tethered to Macs; they would have to work on their own. The stress was getting to some of the engineers. Relationships strained as people spent over 16 hours every day at work. It was too much for one software engineer, Ko Isono, who took his life on December 12, 1992, three weeks before the scheduled CES demonstration.

Apple poured resources into Newton. The company hired psychologists to make sure that everybody on the team was mentally healthy and even started a "buddy program" for engineers who were on the verge of burning out.

New employees were being hired all the time. Tchao now had a staff of almost a dozen people preparing press packets stuffed with news releases, photographs, and a cardboard mock up of a MessagePad.

The engineering staff had doubled several times over. All of the engineers who were working on the Senior tablet moved to MessagePad, and Apple was constantly adding people to the team.

In some ways, this actually made the engineers less productive, since they had to explain the inner workings of the enormously complex MessagePad to every new person on the team. Still, with an estimated 750,000 lines of code, the team needed as much help as it could get.

The CES show went well. Apple had reserved a Las Vegas ballroom for the demonstration, which was to include handwriting recognition, faxing, shape recognition, and infrared beaming (something totally unprecedented in the computer world).

Most of the team arrived a day before the demonstration, and the prospects looked bleak. The MessagePads failed sporadically and without warning during their practices. The power manager was still not totally functional, so the MessagePads would eat batteries constantly. Michael Tchao was dejected by the chances of a successful show.

The demonstration went flawlessly until the very end. The climax of the demonstration was supposed to be two Newtons beaming notes between each other, but they failed. Tchao panicked and rushed into the crowd with Capps to show the gathered reporters the beam function immediately after the engineers left the stage.

It worked this time.

Some reporters grumbled about the changing ship date, but most were impressed, and the coverage was resoundingly positive. Nonetheless, the morale of the Newton team was terrible. Tesler, who had championed the project since 1987, left Newton to become Chief Scientist at ATG.

Spring 1993 was a time of transition for the Newton. It had outgrown its beloved offices at Bubb Road (especially since the Pink project occupied half of it), so Newton moved to the brand new research and development headquarters on 1 Infinite Loop. Not only did they have to contend with a move to new offices, but the team had to give a demonstration at the largest technology expo in the world, CeBit, which is held annually in Hanover.

Over 500,000 people flood into the mid-sized city, forcing event organizers to turn the autobahn leading to the convention center into a one way road to get people to and from the show.

Amstrad Pen-Pad

This event was important not only to show the press that progress was being made, but to prove to Sharp that Apple was serious about finishing Newton. As Tchao and his assistant walked past a beer garden in Hanover, they spied a flyer with the heading "First to Market?" Inside they found a description of a product physically very similar to the MessagePad. It was called the Amstrad Pen-Pad 600, and it cost £299 (about US$450) - $350 less than the MessagePad (although the price was not yet public knowledge).

Amstrad was started Essex, a small city in England. The company had released a number of very popular home computers during the early 80s. As IBM PCs (and even Macs) became more and more popular in Britain, Amstrad lost market share. By the early 90s, Amstrad was making the bulk of its money off consumer electronics like satellite receivers and stereo systems.

The Pen-Pad project was started as a way for Amstrad to break into a promising new market, but the product was so flawed that it was discontinued months later. Still, nobody knew that at CeBit, and the Newton team was dejected.

Tchao was somewhat heartened by the news, though. The birth of PDAs meant that other companies felt the market had potential, too. Sculley had tipped his hand almost two years before the MessagePad would be ready, so Tchao wasn't terribly surprised that other companies had been able to catch up - especially when their products were not nearly as good as the MessagePad was going to be.

The biggest announcement from Apple at CeBit was the addition of the enormous Siemens as a Newton licensee. Siemens was going to produce a desktop phone based on the Newton. Apple also announced that Motorola would release a wireless-enabled MessagePad called the Marco, and that Random House would be developing custom content for the MessagePad.

The demonstration of the MessagePad fell apart. The handwriting recognition refused to recognize the phrase "Ring Dietrich". But when the presenter pulled another MessagePad from the dais, the problems were solved. Moments later, the MessagePad sent a fax to the German celebrity, Dietrich, who was about to perform after the Apple show.

Michael Tchao opened up for questions and was immediately hit with questions about the Pen-Pad. An audience member asked how much the MessagePad would cost, and when Tchao responded "under $1,000", another asked, "Will it below £299" in reference to the Pen-Pad.

The questioner kept hassling Tchao, and Bastiaens challenged the man to a bet. If the MessagePad did not ship before the end of summer, he would give up his entire wine cellar, which was worth many thousands of dollars.

Then Michael Spindler took the floor and handled questions about Apple's fluxing management.

Alpha to Beta

When the Newton team returned to California, they were shocked to hear that Newton's software was now in beta, which meant that it should no longer crash. That was far from the case, and it pushed the Newton team even harder. Few engineers respected the change, especially Steve Capps. Everybody was adding bug fixes to the still unstable code.

One of the software managers was flummoxed to find that Capps kept adding features, even after alpha and now during beta. Capps eventually relented and set to work. The real deadline for the software was May 26, which was when Sharp needed the software to burn in the MessagePads on the assembly line.

On the first day of May, there were of 3,700 bugs in the Newton software. As each day went by, more and more of them were squashed until May 26, when it was clear the software was still too buggy to ship. There were over a thousand documented bugs that had yet to be squashed.

Apple implemented creative incentives to get people to fix as many bugs as they could (like gift certificates to Tower Records), but it was no use. Apple had changed the deadline three times now, and Bastiaens knew that the straight laced Japanese would be furious if Apple caused the manufacturing schedule to fall apart.

Bastiaens decided to make the gamble of producing 4,000 MessagePads with beta quality software, not the Gold Master, which was the stage ready for consumers. Those MessagePads that had beta software would not go to consumers; instead they would be used in point of purchase displays at retailers.

Sculley Out, Spindler at the Top

Just as the Newton team was getting ready to release their first product, their greatest champion was forced out. On June 7, 1993, Sculley resigned at the urging of the board after Apple lost $183.5 million during the third quarter. Sculley was immediately replaced by Michael Spindler. Like Jobs before him, Sculley retained the title of chairman, but he had little to do. A few weeks later he moved to Connecticut to join his wife.

Despite setbacks, the Newton team continued to prepare for its final demonstration before the launch at CES Chicago. Apple rented a nightclub to show off the device, which would be vying for attention from consumer electronics giants Casio and Tandy, which released the Palm Computing-developed Zoomer PDA during the show.

The presenters were going through the same demonstrations they had given at CeBit. As reporters crowded into the club, Apple reps spotted several people holding PenPoint tablets, which had been developed by AT&T and released earlier in the year.

The incredulous crowd watched as the Newton team went through the motions and clapped politely. It was clear, however, that Newton's thunder had been stolen, and no amount of money could buy it back.

"You want to know something really depressing?" Capps asked another engineer, "Joey [a young software engineer] was still in high school when we started this project."

After nonstop programming through the rest of June, Gold Master was reached and manufacturing began.

Newton in Production

The Newton team was not home free yet. Several problems cropped up. The first (and most disheartening) was a flaw with the Getting Started card bundled with every MessagePad. One of the games included on the card to help train the handwriting recognition software failed to work. A patch was quickly put together, but it had to be applied by hand. Over two thousand MessagePads had to have their PCMCIA cards replaced (they shipped inside the MessagePad) before they could be shipped to the States.

Another flaw was discovered only when the MessagePads arrived in an Apple warehouse in Sacramento. The rubber pads on the bottom of the units were not attached correctly and would fall off. Vikki Pachera, who helped with manufacturing, drove to Sacramento with a team of friends and spent an entire day gluing 800 "nipples" on the bottom of MessagePads.

Sculley and Tchao were getting ready for a battery of demonstrations just prior to launch, culminating in a demonstration for Good Morning America on July 30. Tchao was terrified, since he would have to demonstrate the fax feature by sending a fax to Des Moines, Iowa, a part of the country prone to phone line failures. To top that, the area was experiencing serious flooding. All attempts to connect to the area before the show were unsuccessful, so it was up in the air whether Tchao would be able to connect period, much less send the fax.

When the show began, Tchao gave the standard demo of the MessagePad's features, and then he started the fax. Spencer Christian was on assignment in Des Moines reporting on the flooding. Christian's fax machine whirred up and then printed out the fax from Tchao. In front of 16 million viewers, the MessagePad had performed perfectly.

Three days later, on August 2, 1993, Macworld Boston began. Thousands of show goers streamed out of Logan International Airport and got into cabs with Apple ads on the top featuring a photograph of the MessagePad and the text "NOW TAKING ORDERS".

The debut had a campy feel, based on Paul Revere's ride. A brief promotional video was put together (one of the last ever produced by the AppleTV division in Cupertino) that featured the line "Newton is coming! Newton is coming!"

A line of dignitaries gave brief speeches - even Tom Selleck and Tom Clancy appeared to talk about how the MessagePad would affect their Kyle Foundation, a charity for ill children.

As showgoers left the auditorium, they walked past the point of purchase displays that were by now in stores across America, each one with a real Newton inside (running beta software) with demo software finalized by Capps the night before.

Newton's launch was successful, and it was up to the rest of the world to decide whether it was a technological breakthrough or dud.

Originally published on Low End Mac, reprinted with permission from Tom Hormby.