As Porat says, ‘Now we have to finish it’

General Magic ended more than two years of speculation on Feb. 8 when it finally and officially revealed its corporate strategy and alliances at the Hotel Macklowe in New York. No sooner had the curtain of secrecy been lifted, however, than it was replaced with a new — and in the end, far more critical — anxiety. As president and CEO Marc Porat says, somewhat ruefully, “Now we have to finish it.” We might add, “And it has to work.”

A quick recap. Based in Mountain View, CA, General Magic spun out from Apple Computer’s Advanced Technology Group in mid-1990. Cofounded by Marc Porat, Bill Atkinson and Andy Hertzfeld, its mission is to develop technologies that transform communication into a fundamental activity — the technological equivalent of breathing and paying taxes — for networks, information services, classical application software and electronic devices ranging from computers to handheld organizers to telephones and other gadgets that haven’t been invented yet.

These technologies will be licensed to hardware manufacturers, software publishers and service providers, thus (hopefully) providing a ubiquitous platform for an even broader range of developers and information providers to create even more new services, communicating applications and hardware products.

REVOLUTIONS ARE MADE BY DANCING WITH THE BEARS

The folks at General Magic completely believe that the technologies they’re inventing will do nothing less than revolutionize human communication in the coming century. Even to consider achieving that goal, they’ve had to learn to “dance with the bears” — an alliance of powerful corporate partners that they hope will quickly turn Magic’s new technologies into standards.

The bears they’ve selected are pretty good dancers, too. Equity partners to date include Apple, AT&T (the services group, not the products group, which has been rather promiscuous of late), Motorola, and the Holy Trinity of the consumer electronics industry — Sony, Philips and Matsushita. Two of the world’s best known information providers, Mead and News Corp., are also allied with General Magic, though not as equity partners.

But those who read the news stories or attended General Magic’s New York City “coming out” were frustrated by what seemed to be an announcement of no substance, presented in little or no context. There was no demonstration of the two Magic technologies — Magic Cap and Telescript — and nothing was said onstage that surprised anyone who’d been following the rampant pre-announcement speculation in the trade press.

Post-announcement fu. And much to Magic’s chagrin, the day following its big New York debut both the Wall Street Journal and the New York Times reported unnamed AT&T officials as saying the company had “exclusive rights” to Telescript for more than two years. Although AT&T issued a statement to the contrary the next day, the damage was done. Once again, it appeared as though another so-called “open system alliance” had shot itself in the foot.

So what’s the real story? Do the partners have any real interest in seeing the technology succeed, or are alliance members just hedging their bets on the latest buzzword? What does their equity investment buy them? Is General Magic’s technology truly available to all comers? Is the technology real? And just exactly what is the technology? General Magic itself was long on metaphor and short on evidence, so what you’ll read below is our best shot at answering these questions given the information constraints we were under.

CLEARING UP THE CONFUSION

We’re pretty clear on what’s happening with the alliance (more on that later), but progress reports on the technology are still a matter for speculation. Despite repeated requests and numerous interviews both offsite and at General Magic’s Mountain View, CA, headquarters, no demonstrations of Telescript or Magic Cap were to be had.

A leap of faith. Founders told us they didn’t give a public demonstration at the February alliance launch for “competitive reasons” (read: they didn’t want Microsoft to see it), and because they wanted to have it closer to being finished before they showed it off. (Later we were told that there had been a “sneak preview” of Magic Cap in the video shown at the announcement, but that fact, it seems, did not come home to roost with the audience.)

In later, private interviews, they continued to shy away because they said those working on the project were too busy trying to finish it to give demos. Though the reasoning seems slightly disingenuous, for purposes of this article (and because we’ve talked to some of the few who’ve seen the technology in action) we’re taking the leap of faith that the technology actually exists, and is functional to some extent.

THE DIFFERENCE BETWEEN MAGIC CAP AND TELESCRIPT

That said, the most visible technology under development at General Magic is called Magic “Cap” for “Communicating Application Platform,” and that’s exactly what it is. It is a platform — operating system independent, with a customizable user interface — specifically designed for communicating applications and information services.

It was an early version of the Magic Cap platform, actually, that was the genesis of General Magic. Porat, Atkinson and Hertzfeld (as a contractor) were working on a project deep inside Apple to move the company beyond traditional personal computing. As it became clear that such a project would require collaboration between computing, consumer electronics, communications and information industries, Apple decided to spin the company out of Apple as an independent corporation.

Some cake required. As their plan took shape, however, it became clear that Magic Cap was the icing, and that some cake was required. A very substantive piece of technology, the piece that actually enabled the required “personal intelligent communications” to occur, needed to be invented. That’s when Jim White and Rich Miller, both mail and messaging specialists, joined the team and the development of Telescript began.

No one could have been more perfect for the job than Jim White. White invented what’s known as the “remote procedure call,” or RPC, while working at Xerox. (RPC allows a user to contact and conduct transactions on remote computer via a network.) He also led the development effort on X.400 and was a leading contributor to X.500, two worldwide computer mail and messaging protocols.

He and Miller, a former chairman of the Electronic Mail Association, had formed a consultancy for messaging systems and services called Rapport Communication. When Porat started looking around for the sine qua non of engineering talent, White’s name was on many people’s short lists and over to Magic they went.

SAVED FROM THE GHETTO OF INCOMPATIBLE DEVICES

Their work on Telescript began while Porat’s alliance-building continued apace. But at some point a decision was made that, whether they realized it or not, saved General Magic from being banished to the Ghetto of Incompatible Devices.

When AT&T’s Bob Kavner first hooked up with Magic, he said, AT&T was in the process of spending $10 million to port Go Corp.’s PenPoint operating system (platform for the Eo personal communicator) to AT&T’s Hobbit microprocessor. It didn’t seem like a good idea to him to give up the promised benefits of Telescript just because he didn’t want to use Magic Cap as a platform.

Decoupling was critical. “We thought it would be helpful if the operating system was different from the [communication] language,” Kavner said in an interview with Digital Media. “We knew Telescript was hot, but we also knew that interoperability was an important aspect” of the project.

Although General Magic doesn’t make a point of it, it is obvious to those of us outside the compound that the complete decoupling of Magic Cap from Telescript was absolutely critical to the company’s success — especially since Magic Cap can wear many different, well, hats.

How they differ. Since there seems to be a lot of confusion about how and where Telescript is different from Magic Cap, here’s the official explanation. “Magic Cap is a platform for consumer electronics companies who want to create communication devices, and for developers who want to create communicating software,” says Joanna Hoffman, VP of marketing for General Magic and evangelist for Magic Cap and third-party information providers. “In some incarnations, it will be the operating system and user interface as well.”

Magic Cap integrates Telescript, she says, but Telescript does not require Magic Cap. “The important point,” says Hoffman, “is that Magic Cap and Telescript are very tightly integrated in one implementation only — the Magic Cap implementation. Telescript was developed separately, independently of Magic Cap, to be platform independent, meaning not just independent of hardware, but of operating systems as well. It is quite portable.”

The perfect approach. This was the only possible way to be successful. In a world crammed to the gunnels with incompatible computer systems and computing devices, no matter how brilliant Telescript turned out to be, it could never have made itself heard above the noise if it was perceived as inseparable from Magic Cap — as yet another operating system/applications environment à la Newton, Zoomer, Modular Windows, Macintosh, Windows, Unix, First Person, ad infinitum.

TELESCRIPT: THE LINGUA FRANCA FOR COMMUNICATION

Thus, Magic Cap became Telescript’s “proof of concept,” so to speak, the application platform that would show off Telescript to its best advantage. And Telescript became General Magic’s core technology — a communications lingua franca, Porat says, inspired by the PostScript page description language that revolutionized the publishing industry. In much the same way that PostScript provides sophisticated imaging functionality to a wide variety of devices, Telescript is a programming language that provides sophisticated communication functionality anywhere it resides.

Porat says Telescript is an “open language that’s standard, scalable, can live everywhere,” and allows everything from a low-end PC to a super-fast server to a handheld communications device to an information service to a PBX to become “intelligent.”

What does ‘intelligent’ mean? Telescript’s power and functionality revolve around the use of software agents that are imbued with the capability to conduct many kinds of activities over a network — activities such as decision making, fetching information, locating network resources, providing monitor and alarm functions — and can operate across a multitude of different platforms such as those described above.

Although the results are sometimes the same, this is a fundamentally different activity from the kind of transactions conducted, for example, by automated stock trading programs. White says the difference is that Telescript is a way of expressing intelligence, rather than embedding it.

Stock programs and the like often have a selection of intelligent features embedded in the program’s structure, thus anticipated by the designers of the service — i.e., “if X stock drops below $1.50 per share, sell it.” With Telescript, the intelligence to make those decisions isn’t built into the program. It comes from outside — from you, the user. You make your agent look specifically for what you want to know in the data at hand.

Here’s White’s example of the difference: Let’s say you have a subscription to the Sabre airline ticketing database that allows you to purchase tickets electronically. If the Sabre database has been brought into the Telescript universe (a process that White says does not require a major database overhaul), it would be possible for Telescript “agents” in a customer’s portable computer or handheld device or even portable telephone not only to purchase tickets electronically, but also to monitor the flight as it prepares for departure.

This function does not exist in the database itself, and isn’t even necessarily in the customer’s device. The intelligence is expressed in the agent, which operates autonomously between the customer’s device and the Sabre network as part of a “communicating application.”

So if you’re in an important meeting and time is precious, the last thing in the world you need is to leave the meeting early only to find out your flight is two hours late. A Telescript agent, directed by you, the user, can monitor Sabre and notify you immediately if there are any notable changes to your flight. Using the stock program metaphor, an entire software application would be required; using Telescript, a single autonomous agent could handle the job.

HOW TO MAKE E-MAIL SMARTER

There are many other ways that Telescript can facilitate communication. Today’s electronic mail, for example, is passive and rather dumb. I send you an E-mail message and if you aren’t sitting at your computer, or you’re out of town, it will sit there forever until you retrieve it, no matter how important it is.

But in a Telescripted world, I can send my colleague a message that’s imbued with intelligence. Wrapped around that message are instructions that say, “If she doesn’t see this message within two hours, send it to her assistant. If her assistant doesn’t get it, send it to her pager, then to her cellular phone.” Conversely, my colleague’s Telescripted E-mail system can say, “I’m mad at Denise, ignore anything she sends me.”

Any computer software application can be made Telescript-aware. For example, any address book, calendar or database program could be set up to send and receive messages automatically. An entire series of transactions could take place without human intervention.

Tales of power. Let’s look at this more closely, because it’s quite interesting and General Magic uses it frequently as an example of the power of Telescript.

White says Telescript is designed to be able to live anywhere that digital data can live, or anywhere that can respond to digital instructions. Any application can be made sensitive to communication, any network can be made to respond to agents. A Telescript application could be a single agent, while a calendar application could have a dozen agents. There could be one Telescript element living in the local network to mediate appointments or meetings, and another for E-mail that mediates message flow.

So, let’s say that I’m a dentist (God forbid) and Stella is one of my patients. My Telescripted database notes that it is time for Stella’s checkup. My agent then contacts Stella’s Telescripted smart phone- or computer-based calendar program and queries her electronic calendar agent for times she might be available. (By the way, Porat thinks this example is insane, but it does demonstrate the function of agents in the system.)

If Stella has given permission on her end for such a transaction to occur, my agent and her agent could actually schedule her dental appointment without either of us having to answer a phone or voice mail, or otherwise devote any mental bandwidth to the task.

And Telescript is smart enough to know what’s on the other end of the line as well. “For example, if I sent a lunch invitation with sound and animation to an old-fashioned PC, it would figure out what to do with it rather than fail the message,” says Porat.

He calls this, in a rare prosaic moment, “smart messaging,” and believes its benefits will be so smashing that it will actually promulgate the use of agents.

A NEW WAY OF LOOKING AT SOFTWARE DEVELOPMENT

Clearly the Telescript environment brings to fruition the concept of “modular” software and the power of object-oriented technology. Why would a software developer write a whole new calendar program when you can write an inordinately clever agent that fixes the stupidities of the one you’ve already got? If Telescript is widely adopted, it’s likely that we’ll see a new area of expertise develop around distributed data communication applications. Batteries of agents will vie for market supremacy, be they airline monitors, newshounds, traffic reporters or mail filters.

Anticipation of the handheld device and wireless market is creating much of the excitement for Telescript applications today, but that won’t happen for a while. In the shorter term, the group that will be best served by expert communicating agents will be businesses that operate via local area and remote networks, or that find they conduct much of their business from the field.

If Telescript performs as promised, it will be able both to increase productivity and to lower the hassle and cost of doing business on a network, whether local or remote. If software developers support Telescript, they may unearth whole new product categories they didn’t even know existed.

“I set out to build a network to not repeat the mistakes I’d been a part of with X.400,” says White of the time he served developing standards by committee. “What we got was custom, intelligent transaction processing, smart messaging, database access, information services and all the rest. It’s big bang for the buck if you think through distribution systems once and for all.”

In addition, Porat says that after the big guys have had their way with General Magic this year and next, his “grass roots lefty vision” leads him to believe that anyone will be able to buy a development kit and start offering information services spontaneously at a very, very low cost of entry.

LET’S HOPE IT’S NOT CONTAGIOUS

Obviously the potential is great, but there are lots of questions to be asked about Magic Cap and Telescript and how they’ll work in the real world. In one meeting, for example, Porat would not even entertain a question about addresses and directories and how you would find people to send them calendar messages, etc. And some work must be done on common data formats between devices, computer systems and telephone networks — a highly non-trivial matter — in order to achieve the goal of true interoperability.

But even more important is the matter of security. Telescript is a programming language that actually has the capability to enter your computer system and change things on your server or hard disk, live and in real time. Of course, it asks permission first. But the last time we checked, sans the permission thing, the concept sounds very much like what we try to prevent computer viruses from doing.

No undocumented features. And even though it would be very difficult to believe that General Magic doesn’t have everyone’s best interests at heart, the truth is that one “undocumented feature,” as bugs are sometimes called, could wreak great havoc not only with many computer systems, but with the credibility and future of General Magic.

White doesn’t go into great detail about the subject of security, but both he and Rich Miller are adamant that the security and integrity of data was uppermost in their minds while designing Telescript. “Security is the most difficult and interesting area for a product like this,” says White. “But do you think AT&T would let software into its network that it didn’t write if it weren’t secure?”

We hope that General Magic’s partners and potential licensees and customers will make Big Noise about security and data protection and privacy. Since Mitch Kapor, cofounder of the Electronic Frontier Foundation, is on Magic’s advisory board, as is Craig McCaw of McCaw Cellular Communications — both of whom are vocal supporters of data protection — we can only hope these issues are being discussed at high levels within all the companies involved. Such discussion is critical as we move into the era of the so-called Data Super Highway, and a language like Telescript could go a long way in properly setting the stage for the future of secure communications.

SPEAKING OF AT&T, WHAT ABOUT THOSE PARTNERS?

None of General Magic’s ambitions could be fulfilled without some fairly powerful partners. We’ve all seen and heard plenty of horror stories about so-called alliances gone awry, and it pays to approach anything that remotely resembles an alliance with caution. Despite the fact that most of General Magic’s partners have been dabbling in a variety of technologies with a variety of companies — making it hard to tell whether they’re really committed, or whether they’re all just covering their bets — Magic has amassed a rather formidable wedge of companies that in a perfect world may give it the jump-start it needs.

Each partner has a limited equity stake and a seat on the board of directors. Porat says what they get from the partnership is “an early look at the technology and a dialog with senior management,” as well as some level of participation in product design discussions.

Porat maintains that no “exclusive” deals were cut with AT&T or any other equity partners, but the fact that there aren’t enough hours in a day to do what needs to be done now means it’s virtually impossible for General Magic to consider any other large, partner-type players for a while. “We have a responsibility to the current alliance members,” says Porat. “We owe it to them to fulfill our obligation, which is to finish it.”

You are what you eat. Porat says 10 companies approached him within a few hours during a recent conference, and back at headquarters the phone does not stop ringing. Among the suitors are Novell, Lotus, Borland, Microsoft and Intel — all the big names in the personal computing world — but Porat says his focus is on getting Telescript and Magic Cap done. “Most small companies die of indigestion, not starvation,” he says.

But considering the many deals all its partners are making in every imaginable business area, are General Magic’s partners just covering their bets? Porat thinks the question overly cynical. “[The partners] have no approval over design, and they don’t have to build a device if they don’t want to,” he said. “But all the CEOs have gone on the record making certain declarations [of commitment]. They’re all talking to us regularly. They know what we’re doing in detail. If you want to be completely cynical you could say that they’re all just splashing money around, but that’s not happening.”

MANY AMBITIOUS PROJECTS UNDER WAY

In fact, all of General Magic’s six “charter” partners have made what appear to be major commitments to Telescript and Magic Cap. Gaston Bastiaens, head of Apple’s Personal Information Electronics division, says Apple’s first Newton product will be shipped without Telescript because it won’t be ready, but it will be ready for the second Newton, and a simple upgrade will be available. Other execs are on the record as saying Telescript will be part of all Apple products eventually.

And although Sony, AT&T and Motorola are pushing the envelope in their Telescript and Magic Cap product development, equally significant is the addition of Philips and Matsushita to the mix, whose reputations as providers of less expensive consumer electronics devices (Sony is still considered the “premium” consumer electronics label) demonstrates some belief that Magic Cap and Telescript will be around long enough to be driven down in cost.

The most ambitious Telescript project, or at least the one that is most enabling to all the others, is the new Telescript-based network that AT&T EasyLink Services is building especially for General Magic’s technologies.

“We’re working very closely with all the companies in the General Magic alliance, as well as with the companies that are licensing Telescript,” says Bill Fallon of AT&T. “The devices [they're making] are designed to work with our services and vice versa.” Basically the only requirement you need to connect to our service is a device with Telescript on it.” Fallon says he doesn’t want to use the term “bundling” to describe how device manufacturers will market their closely coupled products, but he does expect there to be a close relationship between them.

Designed for wireless. Despite the fact that General Magic doesn’t like to lean heavily on the wireless personal communicator as the sole definition for what Telescript is good at — Porat thinks it will take 10 years before such devices even approach ubiquity, and Telescript will be just as valuable for wire-based communication — the fact that Apple, Sony and Motorola are all building communicating PDAs has given AT&T some very interesting ideas about how to shape a network that can handle smart messaging across a wide variety of devices.

The service it is building, according to Fallon, is classified as an Enhanced Network Service, which is in the same category as the voice mail services that the phone company sells today. He calls it “next-generation E-mail,” configured specifically for Telescripted personal communicator users who are often out of touch with their home-base computers.

Fallon says the idea is to build value inside the network using Telescript, instead of moving a “bag of bits” from one device to another. He says today’s electronic mail systems are analogous to sending a fax. “You dial the fax machine and hit the start button — there’s very little enhancement.”

A SLEW OF NEW CAPABILITIES

But the new AT&T “smart messaging” network, he says, will provide significant enhancements to customers, such as authenticating the user, storing messages, and converting messages from one media to another — i.e., you could send a text file into a network and specify that it be converted to fax and sent to one or many fax addresses at once.

AT&T already provides a service called MailTalk that will “read” you your E-mail messages when you’ve made the egregious error of leaving home without your notebook computer. Though he doesn’t specifically include MailTalk in the new product, Fallon says the service AT&T is building “will be multiple media — the principal media will be voice, data and fax or image. Personal communicators, being the very capable devices they are, will be able to deal with any of those media at any particular time. Our service will be able to support them in various ways no matter what media is being used. If a device wants to be doing fax because that makes sense to a customer, for example, it would make a lot of sense for our network service to provide a variety of fax solutions so that people using personal communicators in our network are being well served.”

The art of store and forward. In other words, the new Telescript network raises “store and forward” to an art form, and is really the technology that enables all the “smart messaging” examples we raised earlier, such as the agents in the Sabre database and Stella’s dentist. In fact, the subscription service you’d use to access the Sabre database would likely be resident in the new system.

“One of the general points we’re making is that store and forward or message-passed communication are a critical component of what is emerging as the personal communications story,” says Fallon. “It’s fundamentally different from today’s point-to-point communication, it’s certainly possible with multiple types of devices, and we and General Magic believe that an enhanced network like this will be a critical component in the success factor for customers using personal communicators.”

Despite the fact that the network is still being built, AT&T participated in last month’s announcement because it believes that its enhanced network service will be extremely attractive to information service providers. The scenario is that a PDA user will be able to subscribe to weather, news, traffic, and maybe even game and entertainment services via the new system, and will be able to retrieve information from wherever she happens to be.

“We want the application-development community to know if they have an interest in serving the emerging personal communicator-owning public with whatever applications they have, AT&T will be a good place for them to keep their eye on,” says Fallon. Mead, News Corp. and Intuit are already building services for the Telescript-based service.

SENDING A MESSAGE TO THE PDA COMMUNITY

Perhaps even more importantly, says EasyLink Services president Gordon Bridge, AT&T wanted to make it clear to other PDA manufacturers that it was backing Telescript. It makes good business sense, he says, and it’s the best thing for the customer, who shouldn’t have to worry about different devices that are unable to communicate. “The world is a dangerous place, but Telescript is not a dangerous technology,” says Bridge. “It shouldn’t scare anyone unless you want to create an alternative approach.”

AT&T is clearly betting that having a smart electronic mailbox that can intelligently examine messages sent to you and decide what to do with them will be critical in a future where there are bazillions of messages flying around the network at any given moment. For the foreseeable future of wireless communication, AT&T’s Fallon says, you’ll pay for it and it will certainly be a more expensive proposition than terrestrial communication. But he believes the ability to filter junk messages until you’re at a lower-cost communications facility, and to receive only the messages you specifically request, will be a better use of wireless for most people and will avoid the $400-a-month cellular bill syndrome.

THE DEVICE GUYS ARE HARD AT IT, TOO

After exploring the new EasyLink service with AT&T, you can imagine that Motorola, for example, is unequivocal in its support of General Magic’s technology. Though it is shy about divulging any product details, Pat Richardson, vice president and director of business operations for Motorola Personal Messaging Products (which is part of the paging and wireless data group), said that the personal communicator it is building would be congruent with the information General Magic has given out, based on Motorola’s expertise in wireless data.

Less shy about its product plans, or at least the reach of them, was former Sony Corp. of America president Ron Sommer (who is on his way to Sony Europe to try to work the same magic in a European recession as he has done in United States).

He believes that Telescript devices, such as the one that Sony is building with Magic Cap, could be an excellent front-end to a TV set, and will provide individuals with links to entertainment-based information services. (Porat likes the idea of “subscribing” to games or interactive fiction, where the network delivers one new character a month.)

And Sommer certainly isn’t limiting the device to a certain size or shape. “When something [like a Magic Cap device] is so personal to you, the business is limited only by creativity. You could use such a device to select a program among 500 channels, off cable and direct broadcast satellite, connect to information services, order tickets, or organize your business,” he says. “It could be a personal device or a set-top box. You won’t throw away anything; you won’t get rid of your PC, for example. And you’ll want to do it in a non-obtrusive way. We’re looking at something less complicated than interactive TV — we’re finding what’s missing in personal communications, in a similar way to what was missing before the Walkman or the Discman.”

A PERFECT BRIDGE BETWEEN BUSINESS AND THE CONSUMER

Sony is particularly interested in Magic Cap, Sommer says, because it is a perfect bridge between the business and consumer markets. “The key issue for Sony is to build on our dominance in the personal entertainment market, but it was always clear that at some point in the convergence, something would happen for us in the overlapping zone,” he says. “The overlap is in personal communicating devices — those will be based on Magic Cap and will use Telescript. Magic Cap, we believe, will become the standard platform in the business. For us, the importance of Telescript comes from the Magic Cap window, which allows us to reach consumers and the professional base too.”

Sommer says that Telescript may eventually become part of other Sony personal information and communication products, including its Pyxis geopositioning device, the Data Discman and MMCD players, its extensive line of computer peripheral products and electronics equipment, and perhaps someday the MiniDisc rewritable optical music player recorder. And it will certainly be connected, via some future information service, to Sony Music and Sony Pictures.

“Initially we’re concentrating on MiniDisc to replace the music cassette,” says Sommer. “But it’s not only for music. CD is one of the fastest growing markets for data storage, and some day MD will be a Magic Cap device that will be built into computers.” He says Sony is looking at the issue, but has no idea when such a product would be available, since its first priority is to use MD to expand its personal entertainment markets.

THE QUESTION REMAINS: HOW OPEN IS OPEN?

Porat calls managing the General Magic alliance the company’s “third technology,” and it is certainly proving to be at least as challenging as the other two. Magic Cap, and particularly Telescript, are clearly seductive, and the company’s biggest pain and pleasure today is that many people want to get their hands on it. Porat and Rich Miller, the VP in charge of licensing Telescript, are unequivocal about wanting to make both Telescript and Magic Cap truly open and available.

The perception of exclusivity with the partners, they both say, is okay with them, because it’s a short-term problem of lack of time and energy, as opposed to policy. As Miller says, “Our partners have paid the freight, so they are first in line. We can serve only so many customers and still maintain our quality standards and schedules.”

But after the AT&T public relations fiasco, he warns against thinking that there are any legal restrictions about whom General Magic can talk to and when. “Our decisions regarding licensing are our decisions,” he says. “This is not a puppeteers’ paradise. If [our partners] could pull those strings, they’d have built the products themselves.”

Miller doubts that any new partnership deals will be made before the products are shown this summer, but licensing discussions continue to take place. “The alliance isn’t frozen,” he says. “One benefit of being a company that doesn’t have a lot of venture capitalists looking for a 10x turn on their money makes us that much friendlier. But we’d have to have a seductive offer vis à vis the value of the technology, its extensibility and the bottom line. Given our partners, given their support, we have the luxury of taking the long view.”

Is Microsoft in view? One of the biggest questions is whether that long view will include Microsoft. Both Porat and Miller confirmed that discussions have taken place “at very high levels” within Microsoft about the company supporting Telescript, but wouldn’t say much more. We assume that, as usual, Microsoft executives wanted to see product blueprints before moving forward, and we also assume that General Magic thought it imprudent to do so at this time.

But we also found it interesting that until very recently, the discussion had not reached the ears of Nathan Myhrvold, Microsoft’s highly regarded VP of technology, who is known to appreciate good technology and has no problem with the idea of making deals with other companies. (Myhrvold is quick to point out that “I’m the guy who licensed TrueType [font technology] from Apple at the height of the lawsuit.”)

“I’d like to talk to them,” said Myhrvold, who at the time we talked to him in early March said no one from General Magic has contacted him directly about joining the alliance. (We hear the situation has since been rectified.) “Saying ‘Microsoft will [hurt] us if we don’t show it to them’ becomes self-fulfilling. I’m more than happy to try and entertain suggestions that we be compatible. The market is immature — we’d be willing to give away a little of our edge to make the pie bigger. If there’s no business, then we’re all unhappy campers.”

Miller agrees, and says “there is every reason for me to find a workable and practical way of existing in a world where today Microsoft has absolute lion’s share of platforms that are computers. My job is to make it truly open and available.”

OTHER THAN THAT, MRS. LINCOLN, HOW WAS THE PLAY?

Other than Microsoft, it looks like fairly smooth sailing for General Magic, at least on the Telescript front. Miller says it gets a little stickier arbitrating the areas where Telescript and Magic Cap overlap — the classic problem that Microsoft itself faces. Though the product areas differ slightly, both companies have to convince outside developers that internal groups that compete with third-party developers — at Microsoft, the applications side and at General Magic, the Magic Cap group — don’t get unfair access to core technologies.

Miller says he’s responsible for managing the engineering, business development and marketing for Telescript, and it’s Hoffman’s job to proselytize for Magic Cap, since Magic Cap competes with other PDA platforms. But at this point, obviously, it’s a problem that General Magic will look forward to, since those problems usually arise only after achieving great success.

Doing the dance. Overall, General Magic is having to learn some rather intricate dance steps even to attempt what’s possible with a technology the magnitude of Telescript. Most people roll their eyes when they talk about the company, assuming that it takes an incredible amount of hubris to talk about revolutionizing global communication in one fell swoop. However, if you scratch just below the surface and really listen to what the folks at Magic are saying about what they’re doing, they are actually quite humbled by what they’ve undertaken.

And despite the David-and-Goliath relationship with their partners, they’ve managed to erect some very firm boundaries about how much influence those partners can have. In addition to existing limits on licensing and co-development decisions, Porat says that there’s no chance any of the partners could ever make a go of taking over General Magic because they’re at a standstill — each owns a certain number of shares, but legally can’t buy any more than a certain amount — and may even remain so when the company goes public.

The wild card. Microsoft is certainly a wild card at this point, but it’s unlikely that it will be able to cook up anything even remotely as sophisticated as Telescript on short notice, and equally unlikely that anyone would believe it at this point if it said it did have a Telescript killer. Sources high in the food chain at Sun Microsystems say its new consumer electronics project, called First Person, is not a competitor to Telescript as reported in the New York Times; though the sources would not elaborate, it’s much more likely to be a Magic Cap competitor. But since those are not only expected but already exist, that news is not likely to take much bloom off the rose.

The real bottom line for General Magic, which appears today to have all its ducks nicely in a row, is whether or not the technology will work as promised. This caveat should not be taken lightly, since very few people have actually seen it in action, but considering the parties involved the bet is hardly a long shot. Porat spends a lot of time “level-setting expectations” that the full promise of what Telescript has to offer will take at least 10 years, if not 15 or 20, to reach fruition. He’s promised to have more news this summer; we’re hoping it’s in the form of a fabulous demonstration of Telescript and Magic Cap. But whatever it is, it will have to be impressive enough to keep the momentum going.

Denise Caruso