Goliath and David
David Nation, previously an attorney at Philadelphia firm Drinker Biddle & Reath, served as Bentley’s general counsel between 1995-2015
It was the first big snowstorm of January 1987, dawn was breaking over the heaping drifts, and Keith, Barry, and Ray were rather pleased with themselves. All three had made it to the office at their customary hour—they liked to get started around 6:00 a.m.—beating Lionville’s snowplows. Ray had conquered the slippery roads in his Jeep, Keith and Barry in the brawny new Chevy Blazers they’d recently bought. Now, they gazed out at the daunting snowscape, sipping hot drinks and “patting each other on the back for having the foresight to buy four-wheel-drive vehicles,” as Barry recalled. Nobody, they mused, was going to be getting in with two-wheel-drive today. Who even knew when, or if, a plow would turn up on Gordon Drive.
)
Intergraph ‘originally viewed MicroStation as a way to protect their hardware business.’
Suddenly, their reverie was interrupted by a most peculiar sight. “This green, junked out, rusted Monte Carlo—probably like a 1972— was coming down the street,” Barry recalled. “It had a foot of snow on the roof,” and its driver was peering through a sort of tunnel that had been cut and cleared from part of the windshield. The brothers marveled at this stunning spectacle until Ray broke their puzzlement with an exclamation. “I think maybe that’s George!” he blurted. And sure enough, the aging coupe veered into the parking lot—whereupon it got stuck in a snowdrift, and George Dulchinos scurried up to the office announcing, “I need four guys to help me push my car in!”
Bentley Systems’ newest programmer had just moved from Massachusetts, and he wasn’t going to let a little winter weather keep him from his keyboard.
Dulchinos entered a company whose fortunes had swung decisively since he’d initially accepted the job. Toward the end of 1986, Intergraph invited Bentley Systems to Huntsville to demonstrate MicroStation. The CAD giant had heard enough about the software from its own customers to perhaps wonder whether it had backed the wrong horse in C-CADD. Yet as Keith observed, “They were trying to convince our accounts that they were going to make that software better. So, there was a non-zero probability that we’d end up getting our markets stolen ... or that they would sue us.”
So, down to Alabama they went. Their visit reinforced the power asymmetry between the two companies. From the tony executive lunchroom to the fleet of trucks Intergraph used to deliver their fancy hardware, the Huntsville firm was the picture of a mature incumbent. Nevertheless, its engineering orientation and technical know-how resonated with the Bentley brothers, who found their counterparts to be capable and largely likeable.
“They did some subtle saber-rattling because we were using their file format,” Barry recalled. “We thought we might have the better
of that legal argument, but we also knew that they could bury us in an avalanche of legal maneuverings—so we were polite.”
Ultimately, the parties determined that the shifting contours of the CAD marketplace favored cooperation. “By then there was a bunch of other PC-based CAD systems that were coming out—AutoCAD software, VersaCAD, FastCAD—and Intergraph was starting to hear it from their users,” Barry said. “They really did need a credible competitor, and C-CADD wasn’t it.”
Behind the scenes, Intergraph was also at risk of losing a lucrative client, the Texas Department of Transportation, along with other state DOTs that were fielding pitches from competing workstation vendors like Sun Microsystems and Apollo Computer. According to George Church, who was managing third-party software affiliations for Intergraph at the time and took part in the negotiations, Intergraph “originally viewed MicroStation as a way to protect their hardware business.” The broad strategy, as explicated in the Financial Times later that year, was that if users insisted on buying PC-based systems, Intergraph wanted them to at least choose software that paved the way for a possible upgrade to beefier hardware down the line.1 Church, who later joined Bentley Systems in 1994, also credited Scott Bentley with placing MicroStation directly in Intergraph chief Jim Meadlock’s line of vision. “Scott’s the guy that hustled and got people’s attention, and got orders,” he reflected, singling out the trip to Bayer as a turning point in persuading Meadlock that he couldn’t afford to ignore MicroStation, notwithstanding Intergraph’s new alliance with CNR Research.
So, in January 1987, Intergraph purchased a 50 percent interest in Bentley Systems for $3 million. The deal gave Intergraph exclusive distribution rights to MicroStation, which it pledged to market on both PCs and its own Unix workstation platform. “The division of labor was that we would do all the development,” Barry explained, “and they would do all the marketing, sales, and customer support.”
The significance of this watershed arrangement is hard to overstate. Freed from the burden of marketing and sales, Bentley Systems could devote virtually all its energy to expanding and refining its software. That played to Keith, Barry, and Ray’s strong suits. Plus, keeping up with breakneck advances in computing during the late 1980s was no small task. There were in fact dozens of PC-based CAD companies scrambling for market share by this time, and each new jump in processing power and computer memory gave any one of them a chance to leapfrog the others with new features or faster performance. Any company that rested on its laurels was inviting its own obituary. Nothing preordained Bentley Systems’ longevity, but hindsight suggests that no element was more crucial to the company’s long-term success than
its laser focus on making the software work for every user, no matter what it took. That abiding ethos—which was surely informed by the brothers’ backgrounds in mechanical, electrical, and chemical engineering—would win it loyal and even passionate adherents among the world’s leading infrastructure engineers. And it was cemented by the 1987 strategic alliance with Intergraph, which gave Bentley Systems the security to pursue a user-centric software development culture that became
its lasting hallmark.
We sold exactly 50 percent,” Keith observed. “Nobody does that.
Yet the deal also contained the seeds of conflict that its structure was ill-equipped to resolve. “We sold exactly 50 percent,” Keith observed. “Nobody does that.” Each company got two seats on a reconstituted board of directors—a recipe for deadlock over any meaningful difference of opinion. A 51-49 split would have avoided that, Keith allowed, “but the price was going to be completely different” depending on who got the controlling stake. So, despite the risks it held for both parties, they opted for a 50-50 division. Before the agreement effectively removed Greg from the board, he made a recommendation that would prove immensely significant down the road.
He advised his brothers to hire a lawyer named David Nation to negotiate the terms. Nation, who was then an attorney at the esteemed Philadelphia firm Drinker, Biddle, and Reath, abhorred the 50-50 arrangement. But he was constrained by it. For the Bentley brothers, watching him maneuver within that tight spot was like witnessing a magician slip out of a straitjacket.
“David Nation had this way,” Barry marveled. “First, their lawyer would be disagreeing violently with whatever David Nation was saying. Then, David would just put it in a different way, sounding ever more reasonable—and eventually the guy across the table would be nodding his head and saying, ‘You know, why didn’t we think of that?!’”
Nation succeeded in inserting a crucial provision into the agreement. It stipulated that if Intergraph was ever found to be competing with Bentley Systems, Intergraph would forfeit its exclusive distribution rights to MicroStation, permitting Bentley Systems to sell the software directly. Nation further protected his client’s interests through a stipulation that would grant Bentley Systems an additional board seat in the event of competition, creating a 3-to-2 advantage. “And that,” Keith said, “turned out to have huge, huge implications later on.”
Measured in terms of the royalty checks it generated and the inside-track marketing position it gave Bentley Systems to the most sophisticated CAD users in the engineering and infrastructure space, the Intergraph alliance was a home run. “It went pretty well on the software development front,” said Barry, who regarded the first few years as a honeymoon phase. Within months of tying the knot, MicroStation was making inroads into places like the UK-based architectural practice of Richard Rogers, the iconic designer (with Renzo Piano) of the Centre Pompidou in Paris and the Lloyd’s building in London.2 In 1989 A-E-C Automation Newsletter concluded that users would “probably be amazed at the completeness of MicroStation’s implementation of [Intergraph’s] IGDS,” lauding its “speed, power, and ease of use.”3 Product reviews in publications like PC Week echoed similar themes.
Yet even the earliest innings of the relationship foreshadowed its ultimate end game. As George Church recalled, Intergraph launched MicroStation under its own umbrella at a trade show in the spring of 1987. To all outward appearances, the debut fulfilled Intergraph’s declared intention to market the software on both PC platforms and its own hardware. A keen eye, however, would have discerned a more complicated reality. “We deliberately put it on the trade show floor on slower PCs than were available,” Church said, “because we didn’t want the workstations to look bad” by comparison.
Intergraph had a hardware culture
The markets for CAD grew exponentially
Meanwhile, on the other side of the Atlantic Ocean, a fresh-faced Intergraph France application engineer named JB Monnier was getting an object lesson in his new employer’s priorities. It was delivered at a management meeting where Monnier, whose doctoral studies in aerospace engineering at the Illinois Institute of Technology had initiated him in the possibilities of the PC, found himself at a long table surrounded by classically oriented hardware types. “As the junior guy, I had picked a seat at the end of the table, opposite the door,” he would recall. “The country manager walks into the meeting late, with a package under his arm, and the room turns quiet. He looks at us and says, ‘We just bought ourselves a PC software company,’ and then he turns his head over his shoulder and pretends to spit on the floor. ‘Here is the product,’ he continues.
It’s called MicroStation.’ And as he throws the box sliding along the length of the table, he shouts out, ‘Anyone interested?’ Somehow the box skids all the way to my end and I raise my arm, ‘I’ll take it.’ And then I hear him say, ‘Fine. You got it, JB—but you are on your own. Now, get out of here, and don’t you dare ask for any support or resources!’ And just like that, I was in the thick of an internal revolution.”
:quality(90))
By the time an Auburn University architecture graduate named Tom Anderson took a job at Intergraph in 1989, “it was apparent the MicroStation team was a separatist group.” The 15- or 20-person group he joined was located three-quarters of a mile away from the company’s corporate campus in a building where they were the only Intergraph employees. Intergraph’s investment in the software was apparent—but so was its ambivalence. “There were a lot of applications already built on top of MicroStation,” Anderson recalled. “Yet it was this really kind of enigmatic thing where the executive management at Intergraph only reluctantly accepted it—or it felt that way.” While Intergraph pre-loaded the software onto the Unix systems it sold, for instance, the company assigned MicroStation PC to be sold through its comparatively weak reseller channel rather than its strong direct sales force.
In short, Bentley Systems stood to reap huge rewards from Intergraph’s sheer scale—but at the outset the bigger company’s brass still regarded MicroStation as a defensive play. “Intergraph sold more, even not trying, in that first year, than Bentley probably could have without Intergraph,” Church reckoned. “But that’s not to say that Intergraph sold anywhere near as much as they could have if they had really tried,” he added. “The culture was hardware, hardware, hardware.”
The problem was that demand was shifting ineluctably toward software. In 1987, the market analytics firm Dataquest estimated that seven PC-based CAD systems were sold for every three larger systems.4 The growth opportunity was enormous. Dataquest reckoned that there were still 10 engineers for every CAD station. But competition was ramping up. So anything less than a full-throated marketing partner put Bentley Systems at risk of falling further behind the emerging heavyweight—a fact that was all too clear to one of the next programmers they tried to attract.
The culture was hardware, hardware, hardware.
Jim Bartlett was teaching computer graphics as a graduate assistant at Cornell, where he’d worked on 3D graphics for a tractor stability simulation program, when a recruiter contacted him about a job opportunity in southeast Pennsylvania. Bartlett and his colleagues had surveyed a good dozen PC CAD packages to figure out which one made sense to teach. The answer seemed clear. “We had determined that the CAD war was over,” he recalled. “AutoCAD software had won.” AutoCAD software, which debuted in 1982, was on its ninth release by the fall of 1987. Its parent company, Autodesk, had gone public in 1985 and was now selling its flagship software to the tune of $1 million per week.5 “It was head and shoulders above all the other packages,” Bartlett thought. “I had never even heard of Bentley. And I had never even heard of Intergraph,” which was essentially a mainframe company, he added. So certain was he in his appraisal, in fact, that he told the recruiter not to bother. It was only when the recruiter counseled Bartlett to do it for the sake of getting “interview experience” that he agreed to go.
And so it came to pass that Bartlett found himself standing in an empty hallway in the Lionville office, next to a placard welcoming him by name, wondering how long he’d have to wait for an actual human being to appear. Presently, a tall man materialized, carrying a suit suspended from a coat hanger. After striding right past Bartlett, Steve Knipmeyer whirled around in a double-take and blurted, “Oh, you must be Jim! Well, I was just going to put on my suit, but now that you’ve seen me, I guess we can just get started.”
It didn’t take long for Bartlett to realize that the five-hour drive had been worth it—and not just because he hadn’t donned a suit, either. He was given a demonstration of MicroStation by a Bentley Systems colleague named Keith Little, whose eventual authorship of tutorial books like Personalizing MicroStation and The Survivor’s Guide to MicroStation would endear him to countless users.
“I was blown away,” Bartlett recalled. “I gave him all the hard problems that I knew AutoCAD software could not handle— and MicroStation was doing it easily ... it was just stunning how good it was.”
“I was blown away,” Bartlett recalled. “I gave him all the hard problems that I knew AutoCAD software could not handle—and MicroStation was doing it easily ... it was just stunning how good it was.”
If Little went to place an element only to find that part of it went off the edge of the screen, he could just zoom out in the middle of the operation and continue—instead of canceling it and starting over, as in AutoCAD software. And that was just one of many AutoCAD software frustrations Bentley Systems seemed to have solved. Then, there was the dual-screen view. At Cornell, Bartlett had been asking everyone why they couldn’t put graphics on two screens. “I’d heard no end of people telling me all the reasons why it was silly to ask that, because it was impossible on account of the way the operating system worked in the PC,” he said. “And here I walk into Keith Little’s office and he’s got graphics on two screens— he’s got eight views simultaneously, in three dimensions, showing and drawing lines from one screen to another!
“I’d had people convince me that was impossible. But Barry and George and so on had figured out how to do that—because they didn’t know it was impossible, so they just figured out how to do it,” he said. “And I was like, ‘Holy cow, how come the world hasn’t heard of this? This is so much better than AutoCAD software.’” Some corners of the world would in fact soon begin to take note. “MicroStation is the Cadillac of high-end computer-aided design packages—mature and solid, despite its tender age,” declared InfoWorld in a product comparison pitting MicroStation V3 against AutoCAD 10 software, Cadkey 3.12, and Versacad 5.4.6 Bentley Systems bested those competitors in “drawing tools” and posted the fastest times in four out of five graphical operations. The only real demerit went to technical support, which the publication rated “poor” because “Intergraph prefers that you call the dealer from which you bought MicroStation PC for support”—and their calls either went unreturned or produced little assistance.
But arguably the most consequential differentiator remained Autodesk’s headstart. “Despite the fact that some products have caught up with it, and surpassed it in some areas, AutoCAD has long served as a CAD industry standard,” InfoWorld observed. “As a result, AutoCAD will be the favored file type if you exchange files with other firms.” And that dynamic was compounded by the massive levels of piracy targeting AutoCAD software, which, with one very short-lived exception, did not employ copy-protection. Insofar as Autodesk confronted piracy, it elected for prosecution over prevention. That strategy may have sacrificed substantial revenue, but it did have one advantage.
)
As “the number of pirated copies soared,” noted CAD historian David Weisberg, it “dramatically increased the number of trained users.”7
Bartlett powered deep into the afternoon talking with the Bentley brothers, Knipmeyer, Dulchinos, and Ernie Sasaki, a “graphics guru,” in Barry’s estimation, who’d also worked for Dynamic Solutions and would contribute to MicroStation through the early 1990s until his path diverged. “I had no coffee, no bathroom breaks, no nothing,” Bartlett remembered. “There was no consideration of creature comforts. There was no social discussion. It was just technology all afternoon.”
It was 1988. In a game attempt to get some “interview practice,” Bartlett had found what would be his professional home for the next three decades.
As Scott continued to helm business operations, the other Bentley brothers remained the nucleus of the development team. Each was settling into a distinct role and personal style.
Keith was widely regarded, by his brothers and other colleagues alike, as the visionary. He had a “really good knack for seeing where the puck was going in the industry,” as George Church later put it, “listening to accounts and understanding what their pain points were.” Eyes perpetually fixed on a horizon that his coding chops brought within reach, Keith also projected a down-to-earth humility and plainspoken directness that fostered collegiality and trust. “When someone asks me what I do,” he said in a short video filmed around this time, “I generally say I’m a programmer—not ‘I’m an executive’ or ‘I’m a president.’” That unshakeable focus on technical issues dovetailed with his unassuming demeanor to potent effect. “There may be no one, outside of my wife, that I have trusted more than Keith,” reflected Tom Anderson, who joined Bentley Systems in 1993. “You can take him at face value. Of all the brothers, I was always closest with him, and I think it had to do with his integrity.”
Ray mixed quiet intensity and irreverent wit with an absolute refusal to be cowed by novel programming challenges, which he attacked with a kind of fearless abandon. “Ray,” said Jim Bartlett in dry understatement, “had a different way of doing things than I had seen in computer graphics classes.” His method could be a thing to behold. He might start by paying a desk visit to Ernie Sasaki, who “subscribed to a million magazines,” which were often the best source of new information in a pre-internet age when neither book publishers nor educational institutions could keep up with the frenetic pace of computer advances.
)
IGUG Las Vegas, 1988
In the late 1980s, dropping a single 400-page issue of PC Magazine on your foot was an excellent way to enrich the local orthopedic surgeon, but Sasaki’s tastes also ran to technical journals like Transactions on Graphics. By his own account, Ray would pick something up and read “for three or four pages,” by which time he’d be impatient to bang the first few lines into an empty screen. He was less bothered by the prospect of stumbling into “a spectacular failure”— which happened plenty—than excited to be flinging his head against a new wall.
“For a good portion of our childhood,” his kids Shelby and Casey reflected upon his retirement, “we weren’t really sure what it was that Dad did for a job. To us, it looked like he just typed really fast on a really loud keyboard ... and often would say words that we were told we were not allowed to repeat when something was not going well.”
For all that, “Ray was famous for writing a new subsystem overnight,” said Anderson. “Even when I was at Intergraph, one of the application engineers bet him a steak dinner that he couldn’t do something—and he did it, like the next day, on MicroStation. He was notorious for how productive he was.”
They were kind of like rock stars... They were doing huge demos of MicroStation on PCs and attracting huge crowds from the very first time.
IGUG Las Vegas, 1988
Colleagues describing Ray often invoked a dictum asserting that there are really just two kinds of computer programmers: pioneers and settlers. Ray was an archetypal pioneer, perennially lighting into new territory and breaking virgin ground. That process was heavy on epiphanies, improvisation, and a certain messiness that inevitably followed in their wake—so the art was to surround him with “settlers who could go in” and fill out the code, as Bhupinder Singh, who joined the development team in 1994, reflected. “When you paired them up, the magic happened.”
Barry excelled at tying everything together, often working deep in the digital guts to build the scaffolding that supported everything else. “I consider myself to be a little more of a computer scientist than Raymond,” he reflected. “I have a better understanding of the low-level parts of the computer.” Barry was in his element navigating operating systems, memory management, software portability, and the interface between hardware and software.
“I did a lot of work on our state machine inside MicroStation—collecting the input from the mouse or a tablet or the keyboard, and integrating that and figuring out where things should go on the screen,” he said. “Back in the old days, in order to get the performance you wanted, you really had to do things in assembly language,” he added, so that became his wheelhouse. Yet he also reached into higher-level processes, like the rubber-banding innovation.
Working at so many crucial software intersections created constant opportunities for friction with counterparts manning any one of them. And Barry’s daunting competence surely bore his father’s imprint. His occasionally gruff demeanor intimidated some colleagues—who quickly learned to think all the way through a problem or idea before walking into his office. Yet the relish he brought to combat belied his strength as a consummate integrator. “He was intense to work with,” said Gino Cortesi, who recalled “battling it out” with Barry countless times during his quarter-century at Bentley Systems. “But he really made you think. And sometimes he would come back after the weekend, even after we’d been arguing, and he would say, ‘Alright, I coded up some things that you can try.’”
A close observer might have noticed something else about Barry, especially as he mellowed with age. He was apt to grade himself low on modesty—but he expressed his most fervent admiration for the excellence of his colleagues. That deep collegiality was a trait all the brothers shared. They also gained a reputation for their accessibility, both within Bentley Systems’ walls and beyond them. In short order they became headline attractions at Intergraph’s user group (IGUG) meetings. From Europe to South Africa, but especially at the annual IGUG conference in Huntsville, they took full advantage of Intergraph’s reach.
“They were kind of like rock stars,” said Tom Anderson, remembering his first few IGUG meetings as an Intergraph application engineer. “They were the demo jocks. They were doing huge demos of MicroStation on PCs and attracting huge crowds from the very first time.
“Nobody wanted to talk to me,” Anderson continued. “But when they were there on the trade show floor, everybody wanted to talk to Keith and Barry. And the reason was that in those days they made themselves available to users. They were doing technical support, they were talking to people. They were just remarkably accessible.”
Keith infused the development team with an ethos he sometimes distilled in presentations as Software = Service. “You’ve got to have a good product, but that’s a given,” he explained. “I think, from Bentley Systems’ perspective, if we’re going to convince you of our products, we first have to convince you of our service. So, early on in my career, I spent a great deal of my time on the phone with users.”
Keith Holce, who joined Bentley Systems much later via the acquisition of Rebis, told a story about dialing the company’s support line in the very early days of MicroStation. “Barry Bentley answered the phone,” he recalled. “I got the answer I needed, but then needed to call support again. This time, Ray Bentley answered the call. I remember thinking at the time, What’s the chance of getting two brothers on two different calls?” Ray settled the second issue, but soon enough Holce ran into a third. “And this time, Keith Bentley answered the phone. At that point, I had to ask,
‘Does everybody in support have a last name of Bentley?’ And Keith responded, ‘Yes!’
In truth, the development team was growing.
Dulchinos hit the ground running. He had little choice, because Barry threw down a challenge almost as soon as he walked through the door. Barry had been writing DOS graphics drivers for the flurry of new graphics cards coming on the market, each of which demanded a unique treatment crafted in assembly language, “a much lower-level and harder-to-understand programming language than the procedural languages that we used for everything else,” as Barry observed. “Anyway, George showed up on his first day, and I had printed out a listing of one of the drivers, pointing out some of the exceptionally clever portions. I was sure George was just nodding his head and claiming to understand. At the end of that one day, I said, ‘Well, George, here’s a new graphics card we want to support—perhaps you could work on that.’ I was positive he would have lots of questions and problems, and I’d have to help him debug his first attempt, etc., but in fact, George said, ‘Ok, got it,’ and took over driver development, lock, stock, and barrel without ever asking me a single question. I can’t remember how many graphics cards we eventually supported under DOS, but it was a lot, and all but the first ones were written entirely by George.”
In an era of extremely limited standardization, Dulchinos became “our hardware guy,” as Scott called him. From graphics cards to digitizer tablets to different types of monitors, there was an awful lot that needed bespoke solutions in order to shake hands with the software. “George was the guy who would dive into the bits and bytes and registers, and make it work.” Jim Bartlett started out working on the MicroStation Unix platform but pivoted quickly to helping develop a version tailored for the Macintosh—which “immediately rolled into” MicroStation V4 for DOS. The official timeline of MicroStation releases has a way of obscuring the fine-grained reality of the development process, which Bartlett likened to the “Agile” framework that was later popularized in a famous 2001 software-development manifesto.
“We used to ship in a kind of semi-continuous stream,” Ray explained. “Whoever compiled at the end of the day would just make disks and we’d send them out to customers.” This facilitated the quick release of small improvements—and, occasionally, opportunities for the developers to lean too far out over their skis. In one oft-recounted incident, Steve Knipmeyer shipped a diskette containing a 3D refinement only to realize too late that the upgrade messed up text displays within it. Ray watched in wonder as Steve bolted from his desk exclaiming, “I’m going to chase the FedEx guy!” It turned out to require more than simply dashing up the street. “I drove to the airport,” Knipmeyer recalled, “found the office, and pleaded with them to go through the bins and find that diskette. I would have been mortified. I did not want the customer to have a bad experience.”
:quality(90))
:quality(90))
Propelled by instinct toward an informal methodology that permitted the constant incorporation of new learning, “We were doing Agile before anybody knew what Agile was,” Bartlett said. “It was just the way we worked.”
Nevertheless, the MicroStation Mac version represented a turning point. “It brings an unprecedented selection of 2D and 3D features to the Mac,” gushed Mac User magazine in July 1989. Architectural Record deemed it a “first-rate package” that “redefines the limits of high-end CAD for the Macintosh” and “melds the Macintosh interface seamlessly to the power of one of the most flexible CAD systems available.” Melding the program to the Mac interface proved to be a significant step. Although MicroStation Mac was released into what turned out to be a commercial cul-de-sac—since Apple’s interests quickly flitted in another direction even though the company had encouraged the program’s development—it set the stage for another major advance via its immediate successor, MicroStation V4 for PCs.
Inspired by the Mac’s graphical user interface (GUI), Barry developed an overlapping-windows interface of his own within DOS—almost three full years before Microsoft’s Windows NT decisively converted the rest of the world to that standard. Another breakthrough came courtesy of Keith and John Gooding, whose history with the Bentley brothers went back to Wassam’s Christmas tree lot. Gooding joined Bentley Systems as a programmer in the late 1980s and collaborated closely with Keith on the development of MicroStation Development Language, or MDL, a C-based extension that allowed users and third-party developers to write customized applications within the program.
With its streamlined ability to call up floating toolboxes and other functions in the windowed environment, MDL enabled these applications to match the look and feel of MicroStation—amplifying the software’s value with add-ons tailored to specific disciplines, from digital terrain modeling to roadway engineering to plant design. Yet the file formats remained backward-compatible with previous versions, a consideration that was especially important for users in the infrastructure space, where asset lifecycles are measured in decades, rather than years.
Gooding effectively created a “virtual machine and virtual operating system that we could run our own applications within,” said Bartlett, who became the core graphics manager for MicroStation. “So we basically had everything we needed to write our own programs, with our own GUI, within our own portable P-code system that we could then port to any platform,” he said. And that innovation permitted Bentley Systems to use the “exact same code base” to make MicroStation run on 14 different platforms.
)
)
)
“That was what
I enjoyed the most,
what was most important
to me: the people.”
The industry was growing so fast that market analysts could barely keep up. In 1987, Computer Graphics World noted that the US stand-alone graphics market was predicted to rise from $8.5 billion that year to over $18 million in 1992. By the time 1992 rolled around and the publication weighed in with the latest forecast, the market had already passed the $40 billion mark.12
The office on Gordon Drive was a beehive of never-ending work. “I remember working from 6:30 in the morning to 6:30 at night, six days a week, practically, and loving it,” said Scott Bentley. “All of us did. We didn’t have much of a life in the late ‘80s and early ‘90s, because we just worked all the time. But we loved it. Ray was funny—he’d say, ‘If I won the lottery, and we were losing money at programming, I’d just keep programming until the money was gone.’ They loved programming. I loved running a company. It was great.”
They worked hard but joyfully. “There was a little grassy patch in front of the office,” recalled Maria Yeo, “and when people really needed to let off steam, they would go out and play dodgeball. Then, they’d come back in and get back to work.”
“At lunchtime,” Barry elaborated, “we’d have 20 people out there throwing balls at each other. And we had a basketball and a lowered rim.”
“At one point,” Ray added, “there was some construction going on next to us, and they had these stacks of sheet rock. So, we’d go out there and we’d play King of the Sheet Rock—you’d each stand on a big pile, and Keith and Barry and I would be out there trying to throw the other guy off. We had so much fun those years.”
“We were fortunate that we were able to hire very good people—smart, young, bright, enthusiastic,” said Yeo, who stayed with Bentley Systems for 31 years. “That was what I enjoyed the most, what was most important to me: the people.”
As the development group grew, so did the need for organization. Keith, Barry, and Ray wanted to stay completely focused on programming, and had little interest in managing others. So, Steve Knipmeyer shifted from the nitty-gritty of coding into a managerial role. “He was a team player,” Barry reflected. Even-tempered and uber-organized, Knipmeyer got along well with others. Perhaps most importantly, he was willing to do it. “We needed more structure,” Knipmeyer recalled. “It was clear that we were going to need to develop some formalism. The company kept growing and growing. We needed to hire, and we needed to have some level of process. That was a role I was happy to fill.”
:quality(90))
Toward the end of 1990, the brothers were contacted by another interested party who thought he could help: Greg Bentley. Having reached a hinge moment in his own career path, Greg offered an infusion of entrepreneurial drive that looked attractive as the company was starting to strain against Intergraph’s leash.
“Keith and Raymond and I are technical entrepreneurs,” Barry said. “We are perfectly willing to dig right in with software, but we’re much less oriented towards doing the business-type stuff.” Greg had ideas about international growth and cultivating third-party developers, and had learned a lot about business strategy and organizational leadership over the past decade.
His brothers also knew the likely outcome of bringing him on board. “We knew that if we hired Greg we could eventually end up with him being in charge,” said Barry. “Greg has too much energy, and always gravitates toward being in control.”
That insight was as old as their experience of siblingship. The younger Bentleys grew up behind an oldest brother whose early-childhood aptitude had become the stuff of family lore. There was the time in second grade when the family’s refrigerator went on the fritz, and Tom and Bonnie went searching for the appliance’s manual only to find it on Greg’s shelf. Or the time, not long after, that he read Gone With the Wind, prompting his fourth-grade teacher to observe that he was the only student reading at his instructor’s level. Delaware named Greg a presidential scholar his senior year of high school (each state selects one boy and one girl)—and even though Barry earned the honor the following year, he couldn’t help feeling that he’d been picked partly for the sake of back-to-back brother novelty. No wonder that Keith, comparing himself to Scott, Barry, and especially Greg, rated himself merely a “good but not great” high school student.
Extracurricular pursuits had a way of following the same template. Barry liked to recount the brothers’ employment trajectory at Wassam’s, a chain of variety stores. They started out as stock boys in satellite locations, but somehow Greg caught the eye of management and eventually progressed to the flagship store. Keith and Barry, who carried on as “working stiffs,” watched as Greg seemed to accrue a veritable portfolio of responsibilities. His rise culminated in his appointment as the boss of Wassam’s annual Christmas tree sales campaign—where he quickly installed his brothers as lieutenants.
)
)
“You could make a lot of money in the three or four weeks before Christmas” earning commissions on $12 trees, Barry recalled. But the real dough only started pouring in when Greg spied a risky opportunity that held the chance of a big score. It stemmed from his observation that Wassam’s adult leadership always wanted to show that they’d sold out of trees a few days before Christmas. Yet that never happened—at least not organically.
“So, Greg would go and negotiate with them, and buy the remaining inventory—maybe 100 or 200 trees—for something like $400 or $500,” recalled Barry, who never lost his relish for narrating his brothers’ triumphs. At that late stage, plenty of them were scraggly runts, but now, the brothers all but owned the market for last-minute buyers and discount seekers. “In the last couple of days,” Barry crowed, “we’d probably double how much we made” over the course of the preceding weeks.
Greg carried similarly high ambitions to the University of Pennsylvania, where he set about trying to earn both a bachelor’s degree and an MBA from Wharton within four years. Informed by an undergraduate advisor that this was all but impossible, he made the rare feat even rarer by completing Penn’s general honors program in parallel—while somehow finding the hours to teach a programming class on the side. “And I got in a little trouble,” he recalled, “because at the same time, I was working full-time as a financial analyst.” Hard work was a family-wide inclination, but as countless Bentley Systems colleagues would come to learn over the years, Greg manifested it in terms of sheer bandwidth.
After graduating in 1977, he joined a small software consultancy called the Yardley Group. He was still working there in 1981 when he married Caroline, but true to form, the couple sold Christmas trees to sock away money for their wedding. Two years and one economic recession later, the precarity of their situation was laid bare. After failing to secure a necessary investment, Yardley Group’s founder was forced to lay Greg off.
“It was a tough, tough time,” Greg recalled. Their first child had just been born with Down Syndrome, and although Andrew would grow into a high-functioning individual whose knack for names and faces would inspire deep affection over his long career as a Bentley Systems colleague, his needs as a newborn were intense. “I went on unemployment,” Greg said. “We had just bought a little house in Devon, and my wife was trying to go back to work, but we needed a lot of early intervention.” The pain of losing his job at a moment of maximal vulnerability would shape his own approach to company leadership in profound ways, from a predisposition to reassign or retrain employees whenever possible rather than dismiss them, to a commitment to broad-based equity sharing. Yet in a present moment that demanded a swift pivot, Greg did not flee for safety. Instead, he tapped an unlikely source of credit, convincing Scott to lend him his Sears card so he could purchase a PC and printer. With that capital expenditure—the sole source of financing—he and two colleagues launched Devon Systems. They aimed to create and sell a currency-trading software kit for banks.
Though he contributed on the programming front, Greg meanwhile secured a consulting gig for Merrill Lynch’s debt-strategy group, which provided revenues to support his partners as they coded around the clock in Philadelphia. Commuting daily to New York City on regional rail, he allowed himself one “luxury”: a Sony Walkman. Thusly fortified by the dulcet tones of The Eagles, and blessed with a wide-open market for their novel software, the eldest Bentley brother helped build a business that quickly broke through to a who’s who of New York and London banks. Greg’s short spell collecting unemployment insurance morphed into success with whipsaw speed. “By 1985, we could pay ourselves $1 million a year or something like that,” he said. In 1987 they were acquired by SunGard Data Systems for $20 million, plus another $30 million in earnouts over the next three years.
Having maximized that earnout, Greg faced a decision about what to do next. He felt that the opportunity for Devon Systems, under SunGard’s umbrella, remained as large as when they’d started it. “And I was all for a hell-for-leather effort to grow aggressively,” he said. “But everyone else was tired, it seemed, and wanted more of a corporate life, focusing on consolidating our position” rather than expanding it.
Looking toward Lionville, he saw a hungry company that still bristled with ambition. So he decided to throw in with Keith, Barry, Ray, and Scott. As Greg later put it, “I preferred to have brothers than partners.”
End notes
Dwyer, John. “Survey of Computers In Manufacturing: Difference between PC and workstation blurs,” Financial Times, June 2, 1987.
“Richard Rogers have recently invested in Intergraph’s latest CAD system,” Building Design, Sep. 25, 1987.
A-E-C Automation Newsletter, August/September 1989, p. 19.
Dwyer, John. “Survey of Computers In Manufacturing: Difference between PC and workstation blurs,” Financial Times, June 2, 1987.
Weisberg, David. History of CAD: Autodesk. Shapr3D, 2008. Link
Gile, Jeffrey R. “Product Comparison: Computer-Aided Design,” InfoWorld, July 17, 1989, pp. 62-72.
Weisberg, David. History of CAD: Autodesk. Shapr3D, 2008. Link
Manifesto For Agile Software Development, 2001. Link
Peltz, David L. “Hands-On CAD,” Mac User, vol. 5, issue 7.
Grey, Nigel. “Microstation goes graphic,” Cadcam, vol. 10, issue 2, February 1991.
Gantz, John. “The market at large: analysis by application dictates optimism for computer graphics industry.” Computer Graphics World, vol. 11, no. 3, March 1988, pp. 27.
Gantz, John. “The market at large: the long-term computer graphics market will benefit from several favorable trends.” Computer Graphics World, vol. 16, no. 1, January 1993, pp. 27.
Solomon, David. “The Fabulous Bentley Brothers,” MicroStation Manager magazine, vol. 2, no. 6, 1992.
“I preferred to have brothers than partners.”
Declaration of Independence