The Second Coming Of Albuquerque's Software Mavericks
by Kyle Silfer
"Great things come out of here for no apparent reason." That's Andrew Stone speaking, and he's on a roll. He means Albuquerque, of course, his adopted home town, birthplace of the home computer (see sidebar), but he could just as easily be referring to the fairy-tale expanse of his North Valley ranchito, where I have come to visit him, tape recorder in hand, on this gorgeous, lazy spring day. Here amid the chickens, llamas and succulent vegetables looms his Castle, a surreal, three-story sanctum sanctorum, rigged out with three computers (a Sun, a Pentium and a NeXT box), ethernet workings, high-speed modems and two ergonomic workstations. Stone is a programmer as well as a small-scale farmer; the Castle is his Place of Coding; and today, with the afternoon sun warming the fertile earth and the power of excellent coffee coursing through his veins, he cannot conceal his delight that he is here instead of there--the traffic-snarled land of drone-like coders known to the world as Silicon Valley. "I love to get up pre-dawn, make that huge latté, and while they are on the freeway," he crows, "I've fixed six bugs!"
Andrew Stone, age 41, has reached this enviable position by what you might call a roundabout path. Born in Cincinnati, he has been a voluntary New Mexican since 1974--first as a UNM architecture student (the Castle is his own handiwork), then as a struggling subsistence farmer in Rio Arriba County, then as a general contractor in Albuquerque and finally--incongruously enough--as a software developer.
"I was using a PC for my business in 1984," he explains. "I was convinced that the IBM PC would be really cool--and it was sucky. Trying to do anything on that was like pulling teeth." But then he "fell in love" with the Macintosh and soon began building HyperCard stacks instead of houses. "That was sort of a major turning point for me. It was more fun to hack with computers than to do construction. ... The turnaround cycle is so fast. You just think of something, you code it, and you can try it." And Stone's hacking soon spawned paying gigs. He returned to UNM for a Masters in Computer Science, but, he says, "I was making so much money, I dropped out."
And then in 1989 "the dog-and-pony show of NeXT came to town." Stone remembers it well. "They carried these huge, monster, foot-by-foot-by-foot titanium black cubes and a 17-inch black-and-white monitor. And basically they sold me on the fact that this was going to be the Macintosh of the future." NeXT Inc., a brazen start-up hatched by Apple Computer's ousted co-founder Steve Jobs, was scouting for talent. When their headhunters met Stone and his soon-to-be coding partner Kris Jensen, they liked what they saw and, over dinner, talked them into becoming NeXT developers.
Things were rosy for a while, but the NeXT platform never really took off--at least not with the "shrinkwrap" consumer market. Consequently, Stone Design has spent the past eight years producing "cool apps"--including a robust drawing application (Create) and database program (DataPhile)--for a dwindling population of users. But all that is about to change. Maybe.
A sudden anomaly in the industry has thrust Stone Design's catalog of useful and interesting software from relative obscurity into the national spotlight. The catalyst for this reversal of fortune? On Dec. 20, 1996, Apple Computer Inc. purchased NeXT Software Inc. from Steve Jobs for a tidy $400 million with the intent of incorporating the NeXT's advanced, object-oriented operating system (NeXTStep) into what will become the new Macintosh Operating System (MacOS), code-named Rhapsody. Almost overnight, old-time NeXT developers were in a position to become Rhapsody developers. For Stone, it was like a homecoming.
"At first," he recalls, "NeXT was going to be for students and publishing, just like the Mac--these are my people, this is where my inspiration comes from--and it moved further and further away from that. NeXT got to be a pricier and pricier deal, which only Fortune 500 companies and the military/industrial/entertainment complex could possibly afford." Unlike other NeXT veterans who viewed the merger as an opportunity to sell out quickly while their properties were hot again, Stone was inspired to reclaim his lost egalitarian user base. "I figure, my God, I've given this thing seven or eight years, might as well give it one more throw of the dice."
So Stone hit the ground running and made headlines this past May for quickly porting Create to run on an extremely primitive version of Rhapsody--just in time to demonstrate it for Apple CEO Gilbert Amelio at the Worldwide Macintosh Developers Conference in San Jose. It was undeniably a big win for Stone Design. "They had never gotten (Rhapsody) to run until maybe a week before," he says. "This thing is running with smoke and mirrors, practically. But Create comes up and works like it's supposed to!"
Stone's swift action has gained him early prominence in a fuzzy, shifting field of play. "It's sort of like, Apple doesn't know what to do, so they've got me up on their shoulder, and they're carrying me around: 'Look! He's just a regular guy, but check out these apps!'" If all goes well, in fact, a demo version of Create will ship with the first developer release of Rhapsody. "The Mac market loves to stay on the edge," Stone asserts. "There'll be tons of users getting the developer release. They're going to want to try all the latest. ... And I don't know how you can get better coverage than that."
Unlikely as it may sound, Stone Design isn't the only Albuquerque NeXT development shop that made national waves in the wake of the Apple-NeXT merger. In an ordinary office building on University Boulevard--as esthetically far from the bucolic splendor of Stone's agrarian compound as you can possibly get--are the unassuming offices of ARDI (Abacus Research and Development Inc.), a software firm devoted since 1986 to the perfection of a single, amazing product known as Executor. This clever program performs the implausible (some would say impossible) feat of allowing Macintosh software to run on non-Macintosh hardware.
How does it work? With deceptive simplicity. You launch Executor
on your operating system of choice--DOS, Windows 95, Linux, OS/2
or NeXTStep--and a very Mac-like desktop appears with the familiar
drop-down menus, windows and folders Mac users have come to expect;
you can then install Mac software--no system software required--and
run it as if you were actually using a Macintosh. This process
of imitating the hardware (CPU and ROMs) and system software (MacOS)
is known as "emulation." And when Executor is running
a compatible application, the emulation is impressive indeed--like
a flawlessly-performed magic trick, only much more useful (appropriately,
the splash screen that appears when Executor starts up is a magician
waving his wand over
The rub is that while many Mac programs run very well, many do not, a fact that led MacWorld--so far the only mainstream U.S. magazine to review Executor--to tag the product "just a toy"--which is a bit of an understatement to say the least. The less-than-perfect compatibility that dogs Executor's heels is primarily due to the fact that the program has been built, over 10 years of labor, without access to any of the jealously-guarded secrets of the Macintosh's internal hardware.
Cliff Matthews, age 34, is ARDI's CEO and chief instigator. A former Burqueño (from 1968-72 while his father was stationed at Kirtland Air Force Base), he returned to get his Masters in Computer Science at UNM (unlike Stone he did not drop out), then returned again in 1988 to dig in and work on Executor. His passion for his work becomes palpable when he explains ARDI's peculiar dilemma: "The sad thing is that it's only difficult because we're honest," he says. "If we were dishonest, then we'd just be ripping the ROMs up, disassembling them ... and we'd have something that's more compatible. In the early days, we fully expected that we would go to court with Apple."
Thus, ARDI has created their "virtual Macintosh" using only publicly-available documentation and a theoretical understanding of the Motorola CPUs found under the hood of every Mac. In other words (as one user sagely posted to the Executor newsgroup): "ARDI has had to figure out what's in the Mac ROMs without reverse-engineering them. Try to figure out how to copy a four-cylinder engine by throwing wadded-up Kleenex at the thing and seeing which way they get deflected, and you'll have an idea of how hard their job is."
Hard job or not, ARDI's accomplishments were significant enough to catch the notice of several journalists pontificating just how Apple would manage to get older Mac apps to run in the brave new environment of Rhapsody. Since, theoretically, Executor/NeXT (like Create) would be simple to port to Rhapsody, articles in various publications--including the San Jose Mercury News, MacWeek and PC Magazine--postulated that perhaps the groundbreaking emulation work of ARDI could be brought to bear on the problem. Matthews agreed and put much of ARDI's development activity on hold last January in order to woo Apple.
Did they get anywhere? "Basically, no," Matthews says. "Eventually, we talked to somebody fairly important in a particular division of Apple. But he clearly didn't understand the technology, and his attitude was, 'You have to convince me that this is all worthwhile, but I'm not really going to take the time to learn the technological side of things because I'm up to my ass in alligators.' Well, the guy was up to his ass in alligators, but part of the reason was that very attitude of 'We're only going to be interested in bright, shiny things.'"
Andrew Stone, a longtime friend and colleague of Matthews, offers further illumination on Apple's position. "I collared this guy who's the head of Rhapsody," Stone says. "And I go, 'You have a product here which would allow you--today--to ship on Windows NT and run Macintosh binaries on any Intel box on the planet. And you guys are shying away. Why? Is it religion? Because if it's religion, I'll accept that as an answer.' And then he finally admitted, 'Yeah, it's religion.' I go, 'Good, because you know what? You don't have a technical answer to this.' ... Apple doesn't want the emotional involvement with Cliff. They can't figure out why one guy can do what they can't do."
So after unsuccessfully courting Apple, dusting off their NeXT code and bringing it up to date, ARDI is back to their main focus: Executor for Intel machines--including the existing, commercially-available Executor/DOS and the incipient Executor/Win32 (which will run more efficiently under Windows 95 and NT). While Matthews meets with venture capitalists, seeking the funds necessary for launching the new, improved Executor 3, he is pragmatic about ARDI's short-term goals. "The Windows port is the immediate source of money," he says. "And it's a no-lose source of money. If Apple tanks, we catch the people abandoning Apple. If Apple grows, we actually catch people going in the other direction. There's a lot of stuff that is MacOS-specific that is available only for the Mac."
Nevertheless, ARDI will be ready for Rhapsody. Since Apple has ultimately decided to provide MacOS support itself for the Motorola Power PC chip and to ignore other chips which will be able to run Rhapsody, ARDI will be the only game in town for Rhapsody/Intel. "There's a question of just how viable Rhapsody/Intel is in general," Matthews notes, "but it's no skin off our back, per se, to do the work. I mean, we already did most of the work as part of courting Apple. So what we did is, we took stuff off the back burner, put it on the front burner, let it get warm and cooked, and we've put it back on the back burner."
The challenge now for both Stone Design and ARDI is for them to bring their products to the mass market. This in turn begs the question: Can the little Albuquerque software houses compete with monoliths like Adobe and Quark in the new Rhapsody arena? Andrew Stone is confident that they can. "The reality that all software developers know is that every single application used today is written by a team of two to four people," he says. "Sure, there may be legions of people who test for bugs and write the help manuals and stuff, but apps are written primarily by one or two people. And the rest of the corporation is there just to make the money. ... I think we can run circles around the big software houses."
Cliff Matthews, who has been forced to pit ARDI against Apple instead of working with them, is more cautiously optimistic. "I personally suspect we'll wind up selling out to Apple. The day that we get some funding from somebody else and start hiring a sales force, I think we're going to get a call from Apple saying, 'Uhhhh, let's talk.'"
Regardless of the final outcome, Stone sees all this NeXT/Rhapsody upheaval as confirmation of something he's suspected all along. "It's been said that the corporation is the dominant lifeform on the planet right now," he remarks. "But I would like to add a corollary that it's the dominant lifeform much in the way the dinosaurs were once a dominant lifeform. And if you look at what corporations are doing, almost all their work is being farmed out. ... They're hiring contractors. So there's plenty of room in this model of these giant entities--I think of them as spheres, and, as you know, you can't pack spheres very tightly." Stone gestures wildly in the air; his blood is up, and he has drained his latté mug. "In fact, what the physicists say is true: Most of space is empty. And because of this, I believe there's plenty of room to maneuver around these entities, using your computer and your modem--to be at home and work."
If Stone is right, perhaps Albuquerque--which has thus far served, at best, as a mere way-station (cf. Intuit) or, at worst, as a low-overhead factory site (cf. Intel, Philips) for the expanding computer industry--will at last have a chance to become the Silicon Mesa it never was. Perhaps the space between the spheres is big enough to hold an entire city. Or at least a loose confederacy of cyber-ranchitos, hacking and farming and staying off the freeway as Stone himself has done.
Before my own caffeine buzz wears off, I drive away from the Stone compound in the golden afternoon light, and I can almost visualize this alternate future myself: a thousand such Castles with a thousand such hackers fabricating an immaterial industry in the green and healthy Rio Grande Bioregion, effortlessly producing money from the sky and fruit from the earth. And then I pull out onto the interstate, signaling to enter the left lane as some asshole cuts me off, and like so much smoke the vision is gone.