Retro Tech

Fax on the beach: The story of the audacious, visionary, totally calamitous iPad of the '90s

AT&T's EO Personal Communicator was going to change everything. Too bad nobody cared.

Back in 1993, a series of ads narrated by a post-Magnum P.I. Tom Selleck and directed by a pre-Fight Club David Fincher offered some eerily prescient visions of the future of tech. My favorite of the bunch begins with a boy reading a book on a large CRT monitor. Selleck’s voice chimes in: “Have you ever borrowed a book… from thousands of miles away?”

The boy, wearing a taqiyah — a cap worn by men in Muslim cultures — taps on the screen to hit a button on the menu. The scene shifts to a man driving a car driving during a heavy rainstorm, his significant other by his side, as Selleck asks, “Crossed the country… without stopping for directions?” The car has an in-dash display featuring a map and a 3-D rendering of the road ahead.

Then we’re transported to a beach. A faceless businessman in an Adirondack chair starts writing on a tablet of some kind, using a device that looks like a ballpoint pen. It’s not immediately clear he’s using a computer — until you see the color screen. Selleck again: “Or sent someone a fax … from the beach?” The fax sends. The man gets up and strolls toward the clear-blue water. And Selleck delivers his pitch: “You will. And the company that will bring it to you? AT&T.”

Two out of three of the tech innovations foretold in this 30-second commercial were dead on: Electronic libraries are common, as is in-car GPS navigation. As for the third: Technically, you can now send a fax from the beach — but who’s doing that in the year 2020? But people are sending emails while enjoying the feel of sand between their toes, so we’ll give that one to AT&T, too.

Of course, AT&T wasn’t the company that ended up bringing us most of the tech predicted in the “You Will” ads. But it did bring that tablet device to market. It’s called the EO Personal Communicator 440, and while not the first mass-manufactured tablet computer — that honor goes to the GRiDPad, a device sold by Radio Shack’s corporate parent Tandy — the EO is generally considered one of the first tablets with mobile connectivity. Released by AT&T in 1993, not long after the telecom giant bought a majority stake in its maker EO, it was a tantalizing glance into the future.

Acquiring an EO Personal Communicator these days is not easy. It is essentially impossible to find on eBay. But I recently managed to get my hands on not one, but two of the units, courtesy of a reader of mine. He sent them to me in a huge box, along with dozens of program disks, an in-depth getting started manual, a software developer kit, and an elaborate stand that included a cellular modem and handset for taking phone calls. (He threw in two working units of the Apple Newton, Cupertino’s unsuccessful first foray into handheld computing, and a non-working IBM Simon, widely considered the first smartphone, for good measure.)

I know my way around vintage tech, but the EO struck me as an odd device. It has a portrait-sized dimension, not unlike a modern 11-inch iPad Pro — but at 2.2 pounds and 0.9 inches wide, it’s twice the weight and nearly four times the thickness of that modern-day device. It felt at once familiar and foreign.

I already knew of the EO’s legendary status and the size of the bet AT&T had made on it. (The “You Will” ad campaign alone cost $40 million.) At one point, EO predicted that its product line could sell 100 million units in seven years. It only sold 10,000 before being discontinued in 1994.

I wanted to know why it failed.

The EO shares little DNA with modern technology, ensuring an uphill battle in getting it to work today. The removable nickel-cadmium battery pack looks like a giant vape pen — not exactly a standard battery design. The device, both in its smaller 440 variant (the version I have) and the larger 880, had numerous ports for things like keyboards, modems, and storage. We were still years out from USB making this mess of wires more manageable.

Plus, this was hardly a plug-and-play device: The EO’s Getting Started guide describes a complicated initial boot process. And the passage of time complicated things. The EO’s rubberized outer shell had turned sticky over the years. I had trouble getting the EO’s power brick, featuring a nonstandard connector, to work, so I had to hunt for a replacement on Amazon. Once I finally plugged it in with a working adapter, the green light powered on, but the device didn’t boot. Its LCD screen — monochrome, unlike the color one in the ad — displayed a glitchy error pattern, akin to what you might get if you turn on an original Game Boy without a cartridge inserted.

It’s always a bummer when something you get in the mail doesn’t work. But in its heyday, the EO was an impressive piece of kit, according to the guy who sent it to me, Matt Meinel, who beta-tested it while analyzing pen-computing options for his then-employer, Swiss Bank (now UBS). The device is powered by its own silicon — the AT&T-developed “Hobbit” CPU, which was originally intended for the Newton before Apple switched gears and went with the rising ARM chipset instead. Said to be two to four times as powerful as Intel’s 386 processor, the Hobbit is especially uncommon; AT&T isn’t known for its microprocessors. Despite splashy announcements, the EO was the only commercially released product that used it.

But the PenPoint operating system, which drove the EO, was the star of the show. Based on the use of gestures (handwritten shapes used for commands such as copy, paste, or delete), the operating system, developed by EO’s original parent company GO, was the first sophisticated graphical computer OS that wasn’t built with a keyboard or mouse in mind. The hardware part of the EO’s pen functionality was driven by Wacom, known in the modern day for its drawing tablets. Wacom’s electromagnetic-resonance technology could detect when the pen was touching the screen and precluded the use of batteries.

This work inspired later platforms, like the original Palm Pilot. It also inspired the tech industry at large. During a 1991 demo of PenPoint at the GO Developers Summit in Silicon Valley, its primary architect, Robert Carr, highlighted its many features for nearly an hour. That presentation helped spark a pen-computing craze in the tech industry. By early 1993, there was a dedicated trade show for pen computing, Pen Expo, which drew major companies like IBM and Tandy.

Meinel was a pen-computing enthusiast back then. He felt Swiss Bank’s salespeople could benefit from it, both technically and socially. “At the time, it wasn't that cool to be whipping out a laptop — it was considered rude,” says Meinel, now senior vice president of sales at the big-data firm Levyx. “So having something that laid flat on a desk was much more socially acceptable in a business meeting — just taking notes on it, as opposed to putting a screen up in somebody's face.”

“My penmanship improved like 1,000 percent when I was using the EO.”

He tried numerous pen-based devices, including the IBM Simon and Apple Newton, but said that the EO was so good it spoiled the others for him — even if the EO’s handwriting recognition wasn’t perfect. “Literally, that device did what my third-grade teacher could not make me learn: how to write clearly,” Meinel says. “My penmanship improved like 1,000 percent when I was using the EO.”

Meinel's use case was niche. He worked at a company that invested millions of dollars in new technology in the early 1990s, always looking for an edge. The EO, which itself cost $3,598 for the $2,799 base unit and $799 cellular add-on (a combined $6,409 with inflation today), was destined to remain an obscure curiosity — priced even beyond contemporaries like the Newton and Tandy Zoomer ($699 each, or $1,245 each with inflation).

But one thing the EO had in its favor was AT&T. If the country’s largest telecom company couldn’t get pen computing across the goal line, who could?

In fall 1992, thanks in no small part to GO’s work on PenPoint, pen computing was the hottest thing on the floor at the fall edition of the COMDEX trade show. And EO and GO, the two companies that brought the EO Personal Communicator to life, had one of the biggest stages in Las Vegas that week.

Robert Carr was there, along with EO’s president, Alain Rossman, ready to show off the device. The stage was supplied by AT&T, an investor in both companies. And according to Jerry Kaplan, a GO cofounder, the technologists made the most of their big moment. “Robert’s demo of PenPoint easily overshadowed the prepared speeches and flashy videos presented by a sequence of AT&T executives,” Kaplan writes in Startup, a 1995 book about the rise and fall of GO. “Then Alain Rossmann showed how a fax could be received on the 440 through a wireless cellular connection, to the delight of the crowd.”

But one thing AT&T couldn’t offer was a defense from the prying eyes of the competition, and Microsoft was looking over GO’s shoulder almost from the beginning.

The 1998 book Barbarians Led by Bill Gates, written from Microsoft’s vantage point, describes an encounter between Carr and Microsoft’s staff in the months before his 1991 demo of PenPoint. Carr was looking for applications for the new operating system — and hoping to get Microsoft’s Office suite on the device. Microsoft employees pushed Carr to use Windows for his tablet instead. The meeting ended in an impasse. Soon after, Microsoft started work on a pen-computing variant of Windows and used its market dominance to limit GO’s access to hardware vendors.

All this meant that by the time any hardware was released based on PenPoint, it was already competing with a version of Windows that supported pen input. Years later, Kaplan would suggest that Microsoft’s pursuit of pen tech, combined with its avenue-limiting control over the hardware market, killed GO’s chances.“[PenPoint] was a technology that was essentially strangled in the crib by Microsoft,” Kaplan tells Input.

But even without competition from Microsoft, there were other limiting factors. The EO cost as much as a used car upon its 1993 release, and it wasn’t selling — its target audience was, effectively, large companies like Meinel’s Swiss Bank. Also, by the fall of 1993 the Apple Newton and Tandy Zoomer were on the market. They cost less, were smaller, and did many of the same things.

AT&T eventually took control of EO, which it bought in mid-1993, and GO, which it purchased outright in early 1994. These acquisitions created a confusing corporate structure. The technologies behind the EO were divided between three different subsidiaries — EO for the hardware, GO for the software, and AT&T Microelectronics for the Hobbit architecture. In Startup, Kaplan blames “the hidden cost of decentralized management and interorganizational communication” for destabilizing the technology.

And it got worse after AT&T killed off the Hobbit in 1994. Though it kept EO going for months afterwards, its engineers quickly left as the telecom giant pushed the team away from hardware production. By that July, GO was a goner, and EO was dead, too. A major element of a $40 million ad campaign couldn’t meet its financial targets.

The failure of the EO brought on lawsuits and lingering frustration. Kaplan, in 2005, filed an antitrust suit against Microsoft that was thrown out on statute-of-limitation grounds. But Microsoft’s aping of PenPoint did eventually cost it some legal fees. The company had to pay a settlement to Alcatel-Lucent, the successor company of AT&T’s iconic R&D facility Bell Labs, as the result of a long-running patent lawsuit involving many of GO’s technologies.

Perhaps the only silver lining to the EO’s failure is that PenPoint was successful enough that it was used by tablet makers for nearly a decade after its parent company died.

In his book, Kaplan notes that, despite the management and competitive issues, “the project might have withstood them if we had succeeded in building a useful product at a reasonable price that met a clear market need.” But Kaplan strikes a different tone today. He knows more about how business works — along with how Microsoft worked in the ’90s. “This was a travesty, both from a corporate standpoint, but more importantly from a consumer standpoint,” Kaplan says. “We could have had the iPhone 10 years earlier, at least.”

“We could have had the iPhone 10 years earlier, at least.”

Today, Kaplan posits that perhaps EO and GO went to the market with the wrong hardware. See, EO was working on a smartphone design, code-named Loki, that could have popularized smartphones years before the iPhone. Information about Loki is scarce (it’s barely even mentioned online) but Kaplan argues that PenPoint could have played well to consumers in that context. After all, one of its most popular applications was a crossword game.

“They started with the iPhone, and went to the iPad,” Kaplan says, referring to Apple. “But we were starting with the iPad and moving to the iPhone.”

But by the time the Loki was being developed, the writing was on the wall at AT&T. We never got to see the device, but the gadgets that EO and AT&T did leave behind are nonetheless fascinating. If only the things still worked. I guess I'll never get a chance to send a fax from the beach using an EO. Thanks for nothing, Tom Selleck.