Sponsored by: Every       Work with the Every team to bring AI to your organization  We launched our consulting arm because we were constant

The Misfit Who B uilt the IBM PC

submited by
Style Pass
2024-06-06 11:00:02

Sponsored by: Every Work with the Every team to bring AI to your organization We launched our consulting arm because we were constantly hearing from readers that while they wanted to implement AI throughout their operations, product, and services, they were struggling to. We're now working with select businesses to co-innovate, research, analyze, recommend, implement, and, finally, train their organizations on AI. Interested? Contact us to start a conversation. Contact Every Want to sponsor Every? Click here. Don Estridge literally gifted the entire personal computing market to IBM. When we first saw him in this series, he was competing with Lore Harp to build the first personal computer. We’ve seen that story from her perspective—now it's time for Estridge's. You might think that the man who ushered IBM into the PC age would have been a central figure within the company, but in reality, he worked in the relative backwater of its Boca Raton, Florida office. This is the story of how Estridge successfully navigated complicated corporate politics to build the iconic IBM PC—and his shocking betrayal. It's the latest installment of The Crazy Ones, Gareth Edwards's monthly column about the forgotten men and women who built the future of technology. Subscribe to Every so that you don't miss out.—Kate LeeSubscribeIn a burnished-oak corridor outside the committee room at IBM’s headquarters in August 1980, two engineers pace nervously. Eventually, a door opens. Their boss, Bill Lowe, emerges from the board room next door. Before they can say anything, he smiles and nods. They laugh. They can’t quite believe it. It’s official. IBM is going to try and build a home computer.Bill Lowe kicked off this ambitious project, but he wouldn’t be the person who would finish it. That role would fall to his successor, a humble, cowboy boot-wearing mid-level executive, out of favor and kicking his heels in the IBM corporate backwater of Boca Raton, Florida. He would take Lowe’s project forward, one nobody else in the company wanted. Just 12 months later, on August 15, 1981, a computer would launch that would change the world: the IBM PC.This is the story of Don Estridge, the man who brought the IBM PC to market and changed business and home computing forever. In just five years he created an IBM division that almost nobody else in the company wanted to exist. By 1983, it had seized 70 percent of the microcomputer market and was valued at over $4 billion ($12 billion today). Under Estridge, IBM’s PC division sold over 1 million machines a year, making it the third largest computer manufacturer in the world on its own. This story is based on contemporary accounts in publications such as InfoWorld, PC magazine, Time, and the New York Times, as well as books such as Blue Magic by James Chposky and Ted Leonsis; Big Blues by Paul Carroll; and Fire in the Valley by Michael Swaine and Paul Frieberger.‘Where’s my Apple?’In 1980, no one senior at IBM wanted to build a microcomputer, as home computers were often called, except its CEO Frank Cary. But one man’s will was far from enough to get things done at IBM, even if he was the CEO. In the late seventies, IBM was vast. Known as “Big Blue” to its friends and enemies, the company had almost 350,000 employees, ran so many branches, and operated in so many markets that one commentator at the time described IBM as a country, not a company. It made mainframe and minicomputers that filled rooms—sometimes, whole buildings—and cost vast sums of money. It made business electronics, dominated the global market for typewriters, and made millions selling a wide variety of office supplies. A former employee would later describe it even more accurately: IBM didn’t stand for “International Business Machines.” It stood for “International Business Mafia.”IBM’s huge divisions, spread across the U.S. and the rest of the world, operated like a series of global mafia families. They cooperated when under threat from the outside, and occasionally fought. The heads of those corporate families would regularly come together at the company headquarters in Armonk, New York to form the Management Committee. There, under the watchful eye of IBM’s CEO, they made decisions that would shape IBM’s business empire globally.Cary started at IBM in 1948. He’d been its CEO since 1972. He had helped it dominate the world of mainframe computing, selling machines that could cost tens—or even hundreds—of thousands of dollars. But he also didn’t believe that the company should rest on its laurels. The Apple II personal computer had been selling in enormous numbers since 1977. At these meetings, he had one question for IBM’s divisional capos.“Where’s my Apple?”His divisional heads always had the same answer. Microcomputers—home computing—were a fad. They were low-cost and low-profit. Let others scrabble around in the metaphorical dirt of home computing. The real money was in the markets that IBM’s divisions already dominated—selling vast mainframes and minicomputer systems to large businesses. Cary was even told to buy Atari, which by then had established itself as America’s home video game system of choice. That’s all home computers were good for: gaming.Frank Cary in 1972. Credit: IBM.But Cary didn’t back down. In July 1980, he finally forced a concession out of John Rogers, the head of IBM’s General Products Division (GPD). Rogers agreed to set up a project to create a prototype IBM microcomputer—as long as it didn’t come out of his budget.The answer lay in Boca Raton, Florida. There, lurking near-forgotten in his vast empire of typewriters, photocopiers and other low-level business electronics, was the Entry Level Systems (ELS) Facility, headed up by one of Rogers’s lieutenants, Bill Lowe. It was a leaky, rundown concrete facility that existed primarily to research and prototype basic electronics. Lowe, who was one of the few microcomputer supporters at IBM, was summoned before the Management Committee in July 1980. He was assigned to “Project Chess” and told to create a prototype of a microcomputer. Lowe was given one month to do so, after suggesting to the Committee that one could be assembled from off-the-shelf parts. And he would report directly to Cary. After all, Cary was footing the bill, not Rogers. To pull off this minor miracle of engineering, Lowe assembled a core team known as the dirty dozen (there were 13 of them in total). They were a mix of engineers and planners taken from a number of different divisions. Mostly, they were individuals who struggled to fit in with the risk-averse, rules-heavy way of working within IBM. Bill Snydes, for example, was a talented engineer who had taken the initiative on a failing project and spun off parts of it into a success. His reward for doing so was a reprimand for going outside of IBM’s standard operating practices. Other misfits, like systems engineer Lew Eggebrecht and software engineer Jack Sam, joined for similar reasons—a chance to work on a project with few limitations. The project seemed doomed to fail. Mainframe computers were very different from microcomputers, and nobody at IBM had any real experience with the latter. Even if suitable off-the-shelf parts could be found, there was no guarantee those parts would work well together. They’d also need to program their prototype to do something that proved it would work. The risk that problems would occur was high, and they’d been given little margin to fix those that did. But it would be a fun way to spend a month. And as most of them were out of favor within IBM, they had little to lose.To their own surprise, after a frantic month of development, this team pulled it off. They created a machine that worked. The computer didn’t do much, but it was enough to show the Management Committee."The system would do two things. It would draw an absolutely beautiful picture of a nude lady, and it would show a picture of a rocket ship blasting off the screen,” Bill Snydes confessed later. “We decided to show the Management Committee the rocket ship."This demo, held in August 1980, convinced Cary and the Committee that the work should continue—for now. They told ELS to turn its prototype into a product—and to do everything else needed to take it to market.Lowe was under no illusions about the difficulty of this task. He was convinced that the project was valuable, but Cary still seemed to be the only senior figure within IBM who agreed. The various division heads wanted nothing to do with the project. But as long as the cost (and any eventual blame) sat squarely on the ELS facility in Boca Raton, they didn’t feel a need to kill it. What the Committee did, instead, was impose a deadline for launch: August 1981. It was another impossibly tight deadline, just one year away.Lowe headed back from the headquarters in Armonk, New York to Boca Raton with Snydes and Eggebrecht. There, the dirty dozen began the necessary work to turn their prototype into a real machine. But Lowe, who was seen as a rising star within the company, was already being considered for a new, bigger role elsewhere in IBM. Soon his promotion to that position was confirmed.Lowe believed in the project but didn’t believe he was the right person to do it. Taking it forward would require a unique set of leadership skills. It had to be led by someone who thrived on breaking the rules and who could use the ELS’s newfound position outside the regular financial and management structure to its maximum. Luckily, Lowe knew just the man within IBM for the job—Philip “Don” Estridge.A company manDon Estridge turned 43 in August 1980. He had been an IBM man his entire professional life. He loved being able to tell people he worked there. He would tell people that his blood ran big, and it ran blue.Don Estridge in 1983. Source: IBM.Yet in many ways, Estridge ran as counter to the grain at IBM as the rest of the dirty dozen did. Tall with a mop of well-kept hair, Estridge wore the crisp suit, shirt, and tie that was the standard uniform for an IBM executive. But he coupled that look with a pair of expensive cowboy boots that often drew frowns from more senior executives. He was a humble man who had no time for flattery and a strong distaste for corporate politics. In IBM, this made him unusual. His fellow senior executives would frequently complain that he failed to return their phone calls, and rarely agreed to join golf sessions or long lunches. And his junior staff noted that he made a deliberate effort never to sit at the head of a table. Estridge believed in collegiality, not seniority.Estridge was hugely charismatic, although his personal magnetism was drawn from a different source than that of Steve Jobs at Apple or Adam Osborne at Osborne Computing. They were men who sold their own visions. They could create “a reality distortion field” around themselves (a term coined by Apple’s Bud Tribble about Jobs). They could convince you of the value of their ideas and inspire you to give everything in their service.Estridge was the polar opposite. He listened and supported. His role was to set objectives and provide people with the resources or political cover they needed. This approach inspired a different kind of loyalty from those who worked for him, but it was just as fierce. And it delivered results.Lowe had seen how effective Estridge had been on the IBM Series/1 minicomputer project. The Series/1 was intended to be flexible enough to serve a variety of big business computing needs. It would be a workhorse that businesses would buy in large numbers, making IBM healthy profits on its five-figure price tag. But its development was plagued with problems and project overruns, and the end result was criticized for being underpowered. As a result, the Series/1 project was regarded as a failure within IBM. However, one small part of it was successful. IBM’s sales department had secured a large contract to convert all of State Farm Insurance’s operations to run on Series/1 machines. That part of the project had been led by Don Estridge, who had managed the production of these specific machines, complete with required software. They had been delivered to State Farm on budget and on time.When the Series/1 project ended, Estridge found himself tied to its wider failure—and out of favor. But IBM didn’t fire people who might still be useful. It simply sidelined them. They would be sent somewhere quiet and remote in case the company decided it needed them again. So Estridge was thanked for his work and sent into the Florida wilderness, to kick his heels in the corporate backwater of Boca Raton.Estridge was a self-confessed computer nut. In his early IBM years, he had worked on radar systems for the military and software for NASA’s Apollo programme. Estridge was amazed by the Altair 8080, the world’s first home computer. He never believed that he might one day have a computer at home. And when Apple released the Apple II in 1977, Estridge bought one for himself. During the day, he would help IBM build mainframes and minicomputers. In the evenings, he would tinker with a computer of his own.When Estridge heard about Project Chess, he wanted in. Lowe was happy to recommend him to the Management Committee as his replacement. In the fall of 1980, Estridge took over. He brought his Apple II with him, setting it up in his office at Boca Raton. If he felt that visitors weren’t seeing the value of microcomputers, they would be subject to an extended personal demonstration of what the Apple II could do until they conceded his point.Going it aloneNobody at the ELS would later remember how the machine they were building acquired the nickname “Acorn,” just that it was a sideways nod to “Apple.” Whatever the name, no one outside of the department seemed to care.Lowe had warned Estridge to expect this. In early September, ELS approached the management of IBM’s new assembly plant in Boulder, Colorado. The new plant was significantly under capacity, and Lowe had offered them the chance to build the Acorn. He had received a polite but firm refusal. Boulder’s senior management would rather let their plant sit idle than associate themselves with Project Chess.However, although IBM senior management clearly didn’t want anything to do with the Acorn, plenty of junior staff did. This would become a trend—senior management would first tell Lowe, then Estridge, to go away. Junior staff would be more interested. Those with useful skills would be lured down south to join Estridge’s ever-growing group of misfits in Entry Level Systems.“We got some excellent people out of Boulder,” Bill Snydes confessed later.Estridge and his senior team quickly realized that the only way they could get their home computer to market was if they did everything themselves. Whenever Estridge was told that yet another IBM department or division refused to cooperate, he would fly from Florida to IBM’s corporate headquarters in Armonk, New York. There, he’d petition Frank Cary for more money or people before returning home once again. It was difficult to secure additional resources at IBM—even if upper management believed in what you were doing. Yet, to the astonishment of his senior staff, Don Estridge was almost always successful. His charisma and quiet persuasiveness got him the concessions he needed. At their height, these trips were happening two or three times a week.Finding a marketEstridge was lucky enough to inherit two things from Lowe: a viable design for a mass-market microcomputer and a core team of competent and enthusiastic engineers. It was Estridge, however, who realized that they should enter the small-to-medium business market, not home computing. The design that Lowe’s team had developed was perfect for this. Nothing about their design was pioneering, but it provided enough power, at a low enough price, to run basic small business software, such as word processors and spreadsheets. Historically, IBM had ignored this market because these companies didn’t have the five- or six-figure budgets needed for mainframes or minicomputers. However, they did have enough money to buy a single machine for $5,000 dollars or less. These same business users associated IBM with good things. As the saying went, no one ever got fired for buying an IBM.However, to turn a profit, they would need to sell in volume, which meant making a lot of machines. Estridge’s options for manufacturing support within IBM were limited, so he took a trip to California to meet Lore Harp, CEO of Vector Graphic. Under Harp’s leadership, Vector had carved out a niche for itself making computers for the small business market. The company also had a reputation for quality that matched IBM’s. Estridge toyed with a partnership. He asked Harp to sign an NDA, and then revealed that IBM was considering a small move into this market itself. The discussion was friendly, but Harp was wary. Estridge left with five Vector microcomputers and a vague agreement that they should talk again. While the offer to partner had been genuine, it would never take place. Instead, Harp would find herself rushing to save her business as Estridge filled out his senior team with individuals interested in going up to the challenge of the ELS alone. SubscribeA well-led teamWord spread quickly that Cary had given Estridge and the ELS freedom to operate outside of IBM’s byzantine structures and processes. As operations expanded rapidly in Boca Raton, it became clear that Estridge wasn’t afraid to use this authority.Dan Wilkie, who came on board to lead production, experienced first hand what a difference this freedom made. With Estridge helping to cut through red tape around budgets, staffing, and processes, Wilkie created an assembly facility from scratch in about four months. The previous unofficial record for doing so at IBM stood at three years. He became one of Estridge’s most loyal and effective lieutenants.“With the PC project, I saw the whole pie for the first time,” Wilkie said later. “We saw the costs, we solved the problems ourselves. We lived with the good and the bad. It’s no exaggeration to say that I made more decisions in the first 30 days with that group than I made during my first 14 years with IBM.”Also critical was the recruitment of H.L. Sparks, known to his friends and colleagues as “Sparky.” A born salesman and a rising star within IBM, Sparks first encountered Estridge after the Series/1 project ended. Sparks could see that Estridge was a talented manager. When the Series/1 project had shown signs of failure, others had shifted the consequences elsewhere. Sparks was impressed that Estridge had done the opposite, shielding his team from blame, even though it meant accepting corporate exile to Boca Raton for himself. Then, in December 1980, Sparks received a call from Estridge, who offered him a job.“A job doing what?” Sparks replied.“Don’t ask. It’s a really great deal. Trust me and just say yes.”Sparks didn’t hesitate, abandoning 18 years of climbing the ladder the IBM way to join the project. Estridge put him in charge of marketing and distribution. There was no point in having a machine if users didn’t know about it or have a way to buy it.Sparks contacted the Office Products Division (OPD) within IBM, which was responsible for selling office items—such as typewriters and photocopiers—to businesses. He’d previously worked with an advertising executive in the division named Jim D’Arezzo and had been impressed with his work. Sparks lured D’Arezzo down to Boca Raton, where he was subjected to one of Estridge’s personal Apple II demonstrations. Sparks and Estridge ran D’Arezzo through the market potential they felt the Acorn could tap into.“I saw right away that this thing was fantastic, and I knew that it was a barn-burner as one of the biggest items of all time in the IBM low-end product category,” D’Arezzo said later.He returned to the OPD offices in New York where he met with Richard Young, the division’s president.“Dick, we need this product,” D’Arezzo insisted.“No, we don’t,” Young replied. Young didn’t like new things. And, as the capo of this particular IBM family, his word was final. Conversation over.D’Arezzo was shocked. Sparks and Estridge weren’t. While Estridge provided senior management cover, Sparks did something heretical instead. He ignored the OPD and worked with mass-market stores directly, negotiating retail partnerships for the Acorn with Sears Roebuck and ComputerLand. Both jumped at the opportunity to sell a product from IBM.The TrampSparks suggested that D’Arezzo join them in Boca Raton. Angry and astonished that his division had rejected the Acorn, D’Arezzo agreed. In January 1981, Estridge welcomed D’Arezzo to Boca Raton as the advertising lead for the Acorn. He told D’Arezzo to take a few weeks and come back with a full marketing plan and advertising campaign, which he would bring to the Management Committee for sign-off in July.“I was sitting there taking notes and so I asked what the product was going to be called when it was introduced. He said he didn’t know. I asked how the product was going to be packaged and he said, ‘Packaged?! We don’t even know what its size will be!’“About this time,” D’Arezzo continued, “I started to get a bit apprehensive and I wondered if I should go home and pack my bags and head back north.”D’Arezzo suggested they stop trying to persuade another division to handle their marketing, and work directly with an external advertising agency instead. This wasn’t how things were meant to be done within IBM, but Estridge again smoothed the way. Freed from IBM’s normally restrictive creative shackles, D’Arezzo and the agency came up with a campaign that featured Charlie Chaplin’s depiction of the Tramp in the 1915 film of the same name. In it, the Tramp is confronted by overwhelming giant machines. This was the IBM of old. Afterward, the Tramp would be shown happily using the Acorn. This was the new IBM.Filling the software gapAs Project Chess’s misfit reputation grew, other divisions started sending their more vocal problem children Estridge’s way. One such arrival was Joyce Wren.Wren worked at one of IBM’s smaller offices in Silicon Valley. She was outspoken on the subjects of sourcing external development, and IBM’s slow pace of software development and management. At one point, needing a program written in a hurry, Wren contracted a couple of programmers directly. She paid for them to stay in a hotel for eight weeks while they wrote it. Her reward was a slap on the wrist from the IBM head office in Armonk. The division that should have gotten the work had complained.But Estridge liked Wren’s approach. He formed a new software commissioning arm within the ELS for Wren to lead. Software was critical to the Acorn project. The biggest requirements were a spreadsheet program and a word processor. These two tools, which were popular requests when buying a computer, delivered the most value for small businesses. The makers of VisiCalc agreed to port their software to IBM’s machine, solving the first problem. A distinctly un-IBM solution was required to deal with the second. They paid the creator of EasyWriter to port his word processor from the Apple II to the new IBM machine. EasyWriter’s creator was John Draper, better known to the world as the infamous “phone phreaker” (and proto-hacker) Captain Crunch. Phone phreaking involved using tricks and tools to make free phone calls illegally. Draper achieved notoriety for being one of the best phreakers in the business. He had only just been released from prison for phreaking when he received the request to convert EasyWriter for Project Chess. He was amused to find himself contracted now to IBM, one of the most respectable (and uptight) companies in America. In Boca Raton, it was decided that it was probably best if Armonk never discovered who the writer of their flagship word processor was.Making a deal with MicrosoftThe final piece of the puzzle for Estridge’s machine was the operating system—something IBM didn’t have time to develop itself.As with many key moments in computing history, the process that led to the adoption of MS-DOS as IBM’s main operating system is veiled in myth. The most popular telling pitches it as a straightforward battle between Bill Gates at Microsoft and Gary Kildall at software company Digital Research for the soul of the new machine—a battle Kildall lost by choosing to take a pleasure flight in his personal plane rather than meet with the men from IBM to discuss a deal.The originator of that story is an unreliable narrator—Bill Gates himself. The reality is far more complex.As early as September 1980, Gates was approached by Jack Sams, one of Lowe’s “dirty dozen” engineers, to provide a version of BASIC for IBM’s new machine. Sams and Gates were old school friends. Sams asked Gates for his opinion on what operating system they should use. Gates recommended that they talk to Digital Research, run by Gary Kildall and Dorothy McEwen Kildall.Gates and the Kildalls had an informal agreement not to compete. Microsoft focused on programming languages like BASIC and other software, while Digital Research focused on operating systems, CP/M. Although Gary wasn’t present at the first meeting, there was no need for him to be there—Dorothy handled the business side of Digital Research.CP/M was the leading operating system for business-focused microcomputers at the time. It was regarded as the market leader in this sector, and a lot of software had been written for it already. As a result, Digital Research was in a strong negotiating position. When IBM offered $250,000 for a fixed-cost universal license to use CP/M, the Kildalls refused. Adam Osborne, the charismatic founder of Osborne Computers, had persuaded Gary to give him a license like that only a few months prior for his new microcomputer, the Osborne 1. As sales exploded, the Kildalls were filled with regret. They watched as Osborne reaped the profits due to the fixed cost they’d agreed to up front and decided they would never make the same mistake again.However, back in Boca Raton, the team was growing frustrated with the pace of negotiations with Digital Research. The Kildalls didn’t fully appreciate how tight deadlines were for the Acorn team. When Gates became aware of this disconnect, he sensed an opportunity. After discussing it with his leadership team at Microsoft—Steve Ballmer, Paul Allen and Kazuhiko Nishi,—the men decided it was time to unilaterally end their truce with Digital Research.Gates revealed to Sams that he was aware of another operating system that could be ported to the IBM machine. Sams confirmed that IBM would be interested—but that the company had no desire to get involved in a series of backroom deals or betrayals. If Microsoft did offer them a viable alternative to CP/M, however, they would consider the offer carefully. Emboldened, Gates negotiated the purchase of the QDOS operating system from fellow Seattle software developers at SCP and renamed it MS-DOS. It was promptly offered to IBM as an alternative option to CP/M.There would be more twists and turns—and a number of legal cases—in the MS-DOS story, but this sequence of events would cement Microsoft as Estridge’s preferred partner during the Acorn’s development and lead to MS-DOS becoming one of the dominant operating systems of the era. Unlike the Kildalls, Gates was prepared to turn over Microsoft’s entire operation in service of the Acorn ahead of launch. And that was exactly what Estridge needed.When the new machine was finally announced to the public it came with a choice of three operating systems: MS-DOS, CP/M, or p-System from the University of California. The choice was the result of a settlement between IBM and Digital Research, which was threatening legal action over the events that had led to the creation of MS-DOS and Microsoft’s relationship with IBM. But with MS-DOS priced at $40 in comparison to CP/M’s $240, it was clear to buyers which version of the machine Big Blue felt they should choose.Family mattersBy March 1981, the hardware for the Acorn was ready to go. By June, the software and operating system were finalized. In July, Estridge received permission to take the product to market from the Management Committee. They would officially launch it in August, right on time. Estridge’s management style had delivered. He pushed his staff hard but was careful to walk the fine line between hard work and burnout.Those who worked in Boca Raton during this period remember that Estridge frowned upon late nights in the office unless they were absolutely necessary. He was a devoted family man and had been married to his college girlfriend, Mary Ann. They had four daughters, one of whom they had adopted after her parents died in a car crash while she was young. He believed that work should not come at the expense of family time. To Mary Ann’s amusement, he would sometimes drive back to the office late at night to check that everyone had gone home. If he found people staying too late, he would demand to know why. If they needed more resources, he would try and find a way to provide them. New transfers to Boca Raton discovered that leisure time was for everyone, not just executives.However, as pressure to meet their launch date increased, Estridge did ask more of his senior team. Wilkie and Sparks recall informal Saturday morning conclaves in the office. It had become impossible to find uninterrupted moments to debate and address problems during the week. These unofficial get-togethers were Estridge’s answer. Attendance was not compulsory for anyone—music recitals and Little League games had priority. Nor did they have set durations. What they offered was a chance for the senior team to get together, take a breath, and work through crises. It was perhaps at one of those sessions that they finally decided on a name. Acorn had always been a placeholder. Internally, its official designation was the IBM 5150. At some point people just started calling it the Personal Computer, “PC” for short. It was a name that seemed like a natural fit. On August 12, 1981, IBM revealed the IBM 5150 to the press and the world. At a grand event in a New York City hotel, journalists were given the opportunity to test it out while Estridge and others stood by, available for questions. Estridge confirmed it would go on sale at Sears and ComputerLand stores from August 15 onward.Source: Byte magazine.The coverage the next day was overwhelmingly positive. The machine was quick, reliable, and well built. The software was good and the price point didn’t break the bank. The press had already fallen in love with the PC, a viable contender to the Apple II for small businesses.“It appears that IBM has a better understanding of why the Apple II is successful than Apple,” the New York Times wrote on the day after its announcement.Changing the worldThe launch of the IBM PC changed computing. Prices started at about $1,500 for a basic model running MS-DOS (about $5,000 today). Most people opted for extras, such as a monitor or additional disk drives, that brought the price up to around $3,000. Still, the price was easily in the range of a similarly fitted-out Apple II and cheaper than a Vector. It was well within the price range of most small businesses. And IBM stood for quality and reliability. It was less risky than buying an Apple. “They didn’t know what to buy until they heard about the IBM PC,” Martin Alpert, the founder of computer hardware manufacturer Tecmar, said, “And then they figured they couldn’t go wrong because it was an IBM.”In July 1981, when the PC was announced, Alpert sensed which way the market was likely to go. By the time it had launched, he had already started the process of pivoting Tecmar to focus entirely on the PC. It was a smart move, one made possible by a design choice led by Estridge. Estridge realized that whoever set the new computing standard would be able to dominate the market for some time to come. The easiest way to set that standard wasn’t just to sell machines; it was to let other companies sell parts, software, and even whole computers that would be compatible with your machine. Unlike all of its major rivals—including the Apple II—the IBM PC was built with an open architecture.The PC swept all before it. Within months, demand exceeded production capacity at Boca Raton by roughly 800 percent.“We can only handle so many factors of two,” Estridge told Byte magazine shortly after launch, when asked how work was going on increasing supply. They’d had a gut feeling that their machine would sell well, but not this well.By the end of 1981, they were shipping 13,000 machines a month out of Boca Raton. It still wasn’t enough to meet demand. By the end of 1983, they had sold 750,000 machines and controlled an estimated 70 percent of the entire home and business computing markets. If IBM had spun the ELS off into a separate business at that time, they would have been the third-largest computer manufacturer on the planet.“I think we’re in an era where the public has adopted computing the same way it adopted the automobile.” Estridge told Byte in 1983.That same year, Time didn’t declare a “Man of the Year.” The magazine declared it a “Machine of the Year.” The PC had conquered the world.The original Personal Computer. Source: IBM.SubscribeThe Woodstock of computingThere are few things in business that can change minds faster than profit. So it was within IBM. In less than two months, the ELS went from the ignored and unwanted child of IBM’s computing arm to its favored son. It was a vindication for Estridge and his team. While Estridge’s management had limited the impact of burnout, getting the PC to market had taken long hours and weekend work. Many—Estridge included—had believed that they’d be able to mentally pause post-launch. Instead, they found themselves under even more pressure than before—to increase production as fast as possible and provide direct consumer support in a way that nobody at IBM had ever needed to do before.Securing the resources to achieve these things took even more trips to Armonk. These were just as successful as before, not least because Estridge was no longer dealing with a Management Committee that viewed the IBM PC with suspicion. But more resources also meant more paperwork and more management responsibility, both for Estridge and his senior staff and engineers. People like Snydes or Wilkie now found themselves having to rely on others to solve technical problems as their teams—and responsibilities—grew. They could no longer dive straight in and fix things themselves. It also wasn’t long before Boca Raton was regarded as the place you needed to work if you were an ambitious IBM executive. Suddenly, everyone was a PC evangelist, including many who the original PC team knew had been actively hostile to their efforts before. One member of the original “dirty dozen” later described Boca Raton as IBM’s Woodstock: Everyone would claim to have been there when it started, and the vast majority were lying.The betrayalAs operations at Boca Raton grew exponentially, it became harder and harder for the original ELS staff to maintain their sense of identity. For many, the joy of working there had come from their status as ignored and unloved misfits. Those days seemed to be over. Boca Raton was changing, and not for the better. They felt like they were being reabsorbed and forgotten. The pressure to reintegrate with IBM’s traditional way of operating began to build. It didn’t help that IBM’s “salarymen” culture didn’t enable financial rewards for exceptional achievements.The ELS had invented an entire industry—the PC business, one that IBM and others were now making a lot of money from. Very little of that money filtered back to the men and women of ELS. Not because IBM didn’t recognize what they’d achieved, but because the company had no way of providing major financial bonuses, at least not ones that weren’t tied to promotion into management. Wilkie, for example, couldn’t help but comment on the difference on a visit to Tandon, one of the companies that supplied IBM with disk drives. “When I passed the executive parking lot, all I saw were Rolls-Royces, Mercedes, and Porsches,” he said. “I did consider that, after all, I was in Southern California, but I also remember thinking, ‘I am definitely in the wrong end of this business.’”Estridge struggled to find ways to financially reward the core members of the PC team without promoting them into management positions—positions many of them didn’t want, because they took them further away from the hands-on engineering they loved. While nobody blamed Estridge for the lack of an alternative reward scheme, its absence bred resentment.And then, in September 1982, the resentment imploded.A phone call arrived, from Martin Alpert, the Tecmar CEO. A group of IBM staff members had contacted Tecmar, offering to sell the company commercially sensitive designs and specifications for the PC. Alpert was an honorable, intelligent man. As a major peripheral manufacturer, his relationship with IBM was the cornerstone of his business. He informed IBM of their potential security breach and then went a step further—he offered to assist them in a sting operation.On September 4, 1982, Alpert met with the rogue IBM employees in a hotel at Cleveland Airport. He listened to their pitch. He told them he was interested and they agreed to talk. He was also wearing a wire.A week later, IBM fired William Erdman and Peter Stearns, two newer members of the IBM PC operation for attempting to sell corporate secrets—along with Lew Eggebrecht, one of the original dirty dozen members and forefathers of the PC.The discovery of Eggebrecht’s betrayal sent shock waves through the original members of the PC project. He had been there from the beginning. He was one of them. They believed that his actions were naive rather than malicious. But it brought all the feelings that something had been lost in Boca Raton back to the fore. It wasn’t even about money. They just wanted someone—something—to recognize that what they had achieved as a group had been special.One day shortly after the incident, Wilkie turned up to a meeting with Estridge and noticed that he was wearing a red rosette on his jacket lapel.“What’s that for?” Wilkie asked.Wordlessly, Estridge opened his briefcase and pulled out another red rosette. He handed it to Wilkie.“What’s this mean?” Wilkie asked again, confused.“It means you’re a member of the finest, most professional and most loyal team that’s ever been assembled in the history of IBM,” Estridge told him. “Put the rosette in your lapel and wear it everywhere you go, so people will know who you are and what you belong to. Besides, we all need to stop and smell the roses from time to time.”Wilkie, a man not prone to displays of emotion, began to well up. It was a small, silly thing, but Estridge had somehow managed to distill all of their frustrations into something tangible.“Now state your business, then get out of here,” Estridge told him. “I’ve got more of these things to hand out.”From that point onward, you could always spot a member of the original PC team at Boca Raton. They were the ones proudly wearing one of Estridge’s red rosettes.The final actThe more successful the PC became, the more IBM’s other divisions—and its more ambitious executives —wanted in. Having finally realized the potential profits to be made in the microcomputer industry, IBM’s most senior management were pushing the ELS for more. Estridge pulled together his weary senior team. They agreed that one of three projects looked promising—a cheaper, cut-back version of the PC (the PC Junior); a high-end version of the existing PC, complete with a hard disk (the PC/XT); or an improved version using newer Intel chips (the PC AT).With everyone at Boca Raton already stretched thin, Dan Wilkie asked the question nobody else wanted to ask.“My god, what happens if the Management Committee approves all three programs?”There was a long, drawn-out silence. Eventually, Estridge spoke. “That’s something we’ll worry about if and when we get that order.”As Wilkie had feared, the order was given.The decision to push all three post-PC projects proved to be the final nail in the coffin for what Estridge had built at Boca Raton. They had become too successful. "You couldn't have a business that might account for 25 percent of your revenue not to be tied into the formal system," a member of the Management Committee from this time later explained.Estridge had fought to maintain the independence of the ELS and keep the rest of IBM away. Now, even his supporters appeared to be working against him. The more resources they gave him, the more he was pressured to take on more standard IBM ways of operation to deal with the size of the operation.  Every promotion inserted additional levels of management and bureaucracy between him and the factory floor. His swift decision-making, approachable nature, and ability to see the full picture had always been his strengths. These became significantly harder to wield once layers of secretaries and junior managers asserted control over his schedule.  All three of the PC successor projects needed Estridge’s managerial hands-on touch. But there was only one Estridge. On August 1, 1983, IBM tacitly confirmed to the company that the misfits of Boca Raton were misfits no more. They were so important and successful that the Management Committee had upgraded Entry Level Systems at Boca Raton. The media and press were informed that ELS had been replaced by the Entry Systems Division in Boca Raton. Its president was Don Estridge.They were a fully-fledged IBM family. That brought with it an implicit requirement to behave the IBM way. Like sharks circling, other divisions began to take back responsibility for activities that they had been happy to let the ELS do before. Distribution was returned to the IBM National Distribution Division. Marketing went to the IBM National Marketing Division. Each change was meant to bring stability; instead, it created complexity and established a more conservative and risk-averse way of operating. That brought delays, which would be blamed on the ESD and Estridge, miscast as problems related to any freedom the division retained.In the end, the failure of the PC Junior in the home market triggered Estridge’s final fall. The project, led by Bill Sydnes, suffered a catalog of delays and failures—in part, because Estridge, still used to managing large projects directly rather than stepping back as a senior executive, could never quite relinquish control. Sydnes, for example, recognized that competing for the home computer market meant going up against Atari and Commodore, and facing off against sharp operators like Commodore’s ruthless founder, Jack Tramiel. In order to beat the master of bargain-basement computing, IBM would have to sink down to Tramiel’s level and sell machines in the kind of discount electronics chains that IBM had always shunned. Estridge, too distant from the minutiae of the project to understand this, vetoed Sydnes’s plans.In its wake, Estridge was finally forced to agree to something he had been trying to avoid for almost a year: accept one final promotion. In early 1985, IBM announced that Don Estridge was now the Director of Worldwide Manufacturing, responsible for 116,000 people at 41 IBM plants in 15 countries worldwide.When Estridge announced the news in the giant corporate auditorium at Boca Raton, the assembled members of the ESD gasped. Wilkie, like many of the old guard present, instinctively reached for the red rosette on his lapel. Some of them began to cry. So did Estridge. On paper, it was a promotion. However, it was a role with no real power. They all knew what it meant. Their leader had been quietly benched, just as he had been in 1979 on the Series/1.The rise of the clonesWhat happened next wasn’t hard to predict. IBM failed to remember one of the key things that made Estridge so successful—his humility. The company assumed that once it controlled 70 percent of the PC market, nobody would be able to compete again. Employees labored away at projects like the PC Junior using the same generation of PC technology, rather than focusing on what was happening in the market around them—the rise of the clones.Estridge’s decision to base the PC on an open architecture was a smart one, but it had been built on a bad assumption—that IBM would continue to move fast and innovate. The clones would always follow in its wake. But IBM took its eye off the ball. The development of a true PC successor, the PS/2, which used Intel’s next-generation processor—the more powerful 80286—proceeded at a snail’s pace.Finally, in April 1987, IBM announced the release of the IBM PS/2, built around the “286” (as the 80286 had come to be known). When it launched, it was already a generation of processors behind. Six months earlier, Compaq, an aggressive PC clone that had lured many of Estridge’s former team away, released the Deskpro. It was the first mass-market machine to use Intel’s new 80386 processor—a “386.”By the end of 1987 market share fell from 70 percent down to 20 percent. IBM finally realized its mistake. It was time to bring Don Estridge off the bench and push IBM to the forefront of the PC business once again.Perhaps, in another timeline, that’s what happened. In this one, Don Estridge was no longer there to save the day.The flightAt 7:30 p.m. on Friday, August 2, 1985, Dan Wilkie was enjoying a quiet evening at home when his phone rang. He was the site manager for Boca Raton and the designated contact in case of an emergency. He listened in horror to the voice at the other end of the phone. An hour earlier, Delta Flight 191 had crashed as it approached the runway in Dallas, Texas, and 137 passengers and crew were now dead.After demand for the PC had soared, the ESD opened an additional manufacturing facility in Texas. The Dallas evening flight had become popular with IBM staff heading between the two sites. For the rest of the night, Wilkie stayed by the phone, coordinating with Delta to identify who from Boca Raton had been onboard. He spoke to worried relatives and promised to use the corporate weight of IBM to support them however they needed, logistically or financially. At 11:30 p.m., he received another call. It was from the manager of an IBM sales office in Fort Lauderdale. Wilkie had been fielding calls from various IBM managers all evening, checking on missing staff. He asked the manager who he wanted to check on.“No, It’s Patty Estridge, Don’s daughter,” the manager said, sounding distraught. “She just called me and said her folks were on the plane.”Wilkie recoiled, as if he’d received a punch to the stomach. For a brief moment, he refused to believe it. Estridge had already left Entry Level Systems. He was preparing to move to New York. There was no reason for him to be on a flight between Texas and Boca Raton. But something deep in his soul told him it was true. He hung up the phone and placed his head in his hands. Four hours later, Delta was finally able to confirm it. Don Estridge was dead.The last goodbyeDon Estridge had always been a family man. After news of his benching by IBM got out, in 1985, Steve Jobs made him an unbelievable offer: $1 million to take over from Jobs and run Apple. Estridge refused. IBM was his family. So were his employees. Even when he’d been made the president of an entire IBM division, he’d continued to drive around the buildings at Boca Raton at night to check on his staff.Most of all, he loved his real family—his wife Mary Ann and his four daughters. They were his life. So, before leaving Florida forever for his new role at IBM headquarters in Armonk he decided to take Mary Ann on a long-overdue camping holiday. They decided to travel via Dallas to meet up with their daughter who lived there."I think he went to his grave with the love affair still going on with his wife," Ed Faber, the CEO of ComputerLand, said. "I know that sounds corny, but you just had to see the two of them swooping around the dance floor to understand how in love they were."The funeral service for Don and Mary Ann Estridge was held on August 5, 1985, attended by their friends and family. They were not alone. Over 1,000 people crowded the church and graveyard that day. They had come, en masse, from Boca Raton.The crowd was a sea of black and gray, flecked with the occasional small flash of red from a lapel. Every one of the pall-bearers for Don’s casket also wore the red rosette..Even after the service ended, the crowd seemed reluctant to leave. Leaving meant admitting that Don Estridge was gone. The man who had always stood behind them, championed them within the conservative culture of IBM, and helped change the world forever. He had given so much of himself to them. On some level, they knew they needed to give some part of themselves back to him. They didn’t want him facing the next world alone.Eventually, it was Wilkie who made the first move. Overwhelmed with emotion, his eyes red and swollen with grief, he stepped forward and detached the red rosette from the lapel of his suit jacket. It was the same one Don had given him years before. Leaning down, he gently placed the rosette on the casket.From behind him, he heard someone else moving through the crowd of mourners. Wilkie didn’t turn around. He knew what was coming. The mourner reached past Wilkie and another red rosette was placed gently, next to his own, on Don’s casket.More movement. Another rosette. Then another.Then, after a while, it was done.Wilkie stepped back, not daring to make eye contact with the friends and colleagues who stood alongside him. He turned and slowly walked away. His colleagues did the same. The spell broken, the other mourners started to disperse, too. Within minutes, all that remained was the silence of the graveyard. On the casket of Don Estridge, a collection of red rosettes cast pink shadows in the pale morning light.Gareth Edwards is a digital strategist, writer, and historian. He has worked for startups and corporations in both the UK and U.S. He is an avid collector of old computers, rare books and interviews, and abandoned cats. Follow him on X, Mastodon, and BlueSky.To read more essays like this, subscribe to Every, and follow us on X at @every and on LinkedIn.Join 70,000+ readers of the best tech writing on the internet

Sponsored by: Every Work with the Every team to bring AI to your organization We launched our consulting arm because we were constantly hearing from readers that while they wanted to implement AI throughout their operations, product, and services, they were struggling to. We're now working with select businesses to co-innovate, research, analyze, recommend, implement, and, finally, train their organizations on AI. Interested? Contact us to start a conversation. Contact Every Want to sponsor Every? Click here.

Leave a Comment