Apple Newton

The Apple Newton failed because the first MessagePad, sold on its ability to read handwriting, became a running joke for misreading it, and Palm's smaller, cheaper Pilot took the market Apple had named. Apple announced the end of Newton development in February 1998.

What it was

The Newton was Apple's pen-based handheld computer, sold as the MessagePad. You entered notes, appointments and contacts mainly by writing on the screen with a stylus.

John Sculley, Apple's CEO, coined the term personal digital assistant at the Consumer Electronics Show in Las Vegas on January 7, 1992, and promised PDAs "selling for under $1000" in 1993, Time's Harry McCracken wrote. The Newton was "little more than a rough prototype" then, The New York Times later reported.

On May 29, 1992, Apple showed a Newton at the Summer Consumer Electronics Show in Chicago. Its press release promised devices that would read a user's handwriting and "transform it into text as they write." It said the first one would be available "in early 1993."

What happened

Apple said on July 30, 1993, that the MessagePad would ship on August 2, priced "from $699 to $949, depending on configuration." It went on sale at Macworld Expo in Boston, fourteen months after the Chicago demo, according to Time. Knight-Ridder reported that it was widely viewed as "critical for the troubled company's future."

By December, The New York Times reported, Apple had sold "little more than 50,000" Newtons in four months, and Apple insiders said sales had "slowed to a trickle" after an early burst.

Sculley had left Apple by then, and the Times wrote that the Newton project "contributed to Mr. Sculley's fall."

Apple kept releasing models, up to the MessagePad 2100 and the eMate 300, a keyboard version for students. In May 1997 it spun the Newton off into a separate unit and looked for an investor, Wired reported later.

After Steve Jobs returned, Apple folded the unit back in. On February 27, 1998, it announced that it would discontinue development of the Newton operating system and the products built on it, including the MessagePad 2100 and the eMate 300.

TidBITS noted that Apple had "reabsorbed the short-lived Newton, Inc. spin-off" less than six months earlier.

Why it failed

Apple said ending the Newton fit its plan to put all its software work into the Mac. Reporters blamed the first MessagePad and the rise of Palm.

What Apple said

Apple's announcement quoted Jobs, then its interim CEO, saying the decision was "consistent with our strategy to focus all of our software development resources on extending the Macintosh operating system."

A Newton FAQ linked from the release said Apple had concluded "that the market opportunity targeted by Newton was not aligned with our new business strategy."

On a conference call, chief financial officer Fred Anderson told reporters that the opportunity in "a vertical market like the MessagePad is very small compared to the opportunity for the Mac OS," Wired reported.

Neither the release nor the FAQ mentions handwriting recognition, sales figures or Palm. Pen Computing's David MacNeill called the missing answer to "Why?" the release's "most galling omission."

What reporting showed

In December 1993, John Markoff wrote in The New York Times that the Newton had been "ridiculed for its mistakes in translating handwriting and called unworthy of its price, $699 and up."

He named Sculley as perhaps the main problem, an executive who "promised the computer buyer too much too soon." The Newton's completion, first scheduled for April 1992, slipped to August 1993, he wrote, and it shipped without many planned features, such as wireless communications and some artificial intelligence.

When Apple ended the line, Wired quoted Jupiter Communications analyst Ross Rubin: "Its handwriting recognition was the butt of jokes, and they only sold 80,000 in the first year."

Wired wrote that the MessagePad had been "overwhelmed" by 3Com's PalmPilot and by devices running Microsoft's Windows CE. Citing Dataquest, it put the Newton at 6 percent of the handheld operating system market in the first half of 1997, against 66 percent for the PalmPilot and 20 percent for Windows CE.

What later analysis found

Harry McCracken used a first-generation MessagePad for six weeks for Time in 2012. He agreed with its early admirers, Walt Mossberg among them, that its handwriting recognition was the finest anyone had seen up to then. It "needed to be good enough to be useful," he wrote, "and in its original form, it wasn't."

People remember a week of Doonesbury strips in August 1993 in which a Newton-like PDA's handwriting recognition didn't work, he noted. Later Newtons got much better at it. "That didn't help enough," he wrote.

He also pointed to its size, roughly that of a VHS tape, as Mossberg had noted in his 1993 review. Palm's Pilot went on sale in 1996 and used Graffiti, which had users enter simplified characters one at a time. "Starting at just $299," McCracken wrote, it "became the phenomenon that the Newton was supposed to be but never was."

The Register's Tony Smith wrote in 2013 that the real flaw in early units was "not the recognition code but a memory-handling bug" that made the recognizer lose data unless the MessagePad was rebooted often. He concluded that "it was too early." Built-in cellular, he wrote, would have raised the price beyond what most businesses could justify.

In Pen Computing's June 1998 issue, David MacNeill counted the lack of a shirt-pocket Newton selling for under $400 among Apple's "most heartbreaking mistakes." Steve Capps, an original leader of the Newton group, told him some of his team had wanted a Pilot-size Newton, but the technology wasn't there in 1991. "We should have believed in our own thinking," he said. "Palm didn't make the same mistakes and they deserve their success."

The stage it reached

The Newton never left the introduction stage, which the product life cycle guide describes as the launch period, when awareness is low and sales are small.

It stayed on sale for four and a half years and sold 150,000 to 300,000 units, by Cult of Mac's estimate. The market was "fast-growing," Wired wrote, and the PalmPilot led it.

I count it as a failure, not a decline, because it never had a growth period to decline from. The Humane AI Pin ended in the same stage in 2025, and The Verge's David Pierce ranked it among gadget failures "up there with Quibi, the Fire Phone, and the Apple Newton."

The lesson

I think the Newton's mistake was made in January 1992, when Apple announced a new category on a rough prototype and promised it for the next year.

Everything after that was a race to match the promise. By the Times' account, the team cut features to ship. McCracken found that the feature people judged it on, reading handwriting, wasn't good enough in its original form.

Palm aimed lower, with a pocket-sized device at $299 that asked people to learn a simpler way of writing. I'd copy that.

If your product rests on one hard feature, test it with ordinary users before you announce anything, and cut the promise down to what works.

When Apple launched the first iPhone nine years after ending the Newton, McCracken wrote, the technology finally let it build the kind of device it had wanted in the 1990s.