Computer Chronicles Revisited 23 — Steve Wozniak, Adam Osborne, Lore Harp, and Gene Amdahl

This next Computer Chronicles episode focused squarely on people rather than products. The formal subject is “computer entrepreneurs.” And the four guests are people who were all quite well known in the computer industry during the early 1980s. What’s fascinating, as we’ll see a bit later, is that two of the guests had ventures that each managed to flame out not long after this episode aired.

“I Had Been Working My Whole Life to Build a Certain Type of Computer for Myself.”

Of course, co-host Gary Kildall was himself a well-known computer entrepreneur, having founded Digital Research in 1974. Stewart Cheifet opened the show by asking Kildall about the changes to the “people side” of the computer industry over the decade that followed. Kildall joked he’d traded in his cowboy boots and jeans for a three-piece suit. On a more serious note, he said the biggest change he’d observed was that the industry went from having no products that were competitive to a market where everyone–including IBM–was now producing machines with multi-million dollar budgets. As a result, the industry had become more professional with higher stakes, but also more fun in Kildall’s estimation.

Instead of the usual B-roll this week, we actually get clips from a sit-down interview with Apple co-founder Steve Wozniak at his home. (The female interviewer is not identified.) Wozniak noted that back when he started working on microcomputers in the mid-1970s, there were no interface cards, peripherals, or software for such machines. This made things exciting for people like himself as they were doing things for the first time.

Indeed, Wozniak emphasized that around this period–1974 to 1975–he was not even thinking about what steps were necessary to have a successful product or company. Instead, his motives were purely personal: “I had been working my whole life to build a certain type of computer for myself.” And he built the best one–the Apple I followed by the Apple II–that was doable at the time with the components available. Wozniak emphasized the “freedom” that came with this approach, as opposed to working on a company project where some manager defined what needed to be done.

Wozniak said that now–1984–everywhere he went there were meetings with people in three-piece suits. The computer industry was a large business with a lot of dollars involved. The people who were coming out of school trained to run and manage a business were key now. Consequently, Wozniak said there were very few technically creative products being developed in the microcomputer field.

The interviewer then asked Wozniak if a “garage operation” was still feasible in the small computer industry. Wozniak said he thought that era was over with respect to hardware. It used to be that people could quickly manufacture simple personal computer boards on a small budget and sell them by putting an ad in a hobbyist magazine. Today, that was still possible for software groups. If you came up with a good idea for a software project, there were avenues to find companies that could market it for you or turn it into a product.

Wozniak added that about once per decade, a very large market would come up from zero to make billions of dollars within a few years. It was possible this could happen again in microelectronics and computers, he said. He could see a new group of people coming along who weren’t part of an established platform like Apple and start new businesses. These new garage operations could then become very large companies within a few years. Of course, they’d be shunned by the established players, just as Apple was at first. Wozniak concluded with the observation that when you’re looking at someone else, you can understand how you were perceived back in those days and realize they were right–it was just a quirk of fate that they were wrong.

“There Will Never Be an IBM in Software”

Back in the studio, Adam Osborne and Lore Harp joined Cheifet and Kildall. Osborne was the man behind the eponymous (and bankrupt) Osborne Computer Corporation, which he’d left to start a new venture, Paperback Software International. Kildall reminisced that he and Osborne had worked together with Intel back in 1973. Kildall then asked Osborne if it was still possible to get into the computer industry or if the market had been completely dominated by companies like IBM and Tandy. Was there still a place for a young inventor to make their mark? Osborne replied that it depended on what area of the market you were talking about. If you were going to build a microcomputer, it probably wasn’t a good idea, as there was nothing hi-tech about that market anymore. That space was now a “Clash of the Titans.” Building a computer in this market was like building a washing machine or a refrigerator, Osborne said, in that it was about the economics of volume and reliability.

On the other hand, Osborne said there were more opportunities than ever before in other areas. He analogized it to selling razor blades as opposed to razors. Cheifet asked for clarification–what areas could an entrepreneur make a mark? Osborne said peripherals, software, and other add-ons. He noted you could make a lot of money selling extra cards that fit into a popular computer. As for software, that part of the industry was still in its infancy. Osborne predicted there would be 10-to-100 times more software sold 5 or 6 years from now. Such hyper-growth would create a lot of opportunities.

Cheifet then asked Osborne about who the “Titans” were that he referenced earlier. Osborne said IBM was the main player. He expected AT&T would come into challenge IBM in a big way. He also said at least one of the Japanese computer companies would come in, and together with IBM and AT&T dominate the market.

Cheifet turned to Harp, the CEO of computer manufacturer Vector Graphic, and asked for her assessment. Harp said she largely agreed with Osborne. She said that in her estimation, it would be cost prohibitive for any new computer company to get started today. When she started Vector back in 1976 it was possible to do so with $5,000 in capital (and a lot of sleepless nights). Today that was impossible. A company needed several million dollars in capitalization, and even then that might not be enough money to get critical mass in a short enough period of time to become a contender. But like Osborne, Harp saw room for growth in computer add-ons, such as communications peripherals that linked systems together.

Kildall asked if she meant communications services like the Source or CompuServe. Harp said that was part of it.

Returning to the idea of the “Titans,” Kildall noted that companies like IBM only entered an area of software or hardware when it was proven to be successful. This meant the small “garage shop” operations still had opportunities where the larger companies were not willing to put in the effort right now. Osborne added there were also price considerations. When people were spending thousands of dollars on a computer, for example, they tended to be very careful about what they bought because it was an investment. In contrast, if you’re buying a $50 product–even as much as a couple of hundred dollars–you’re more likely to say “what the heck” and buy it, as even if it doesn’t work out you’ve still learned something. Osborne said that was why you saw a lot of very inexpensive computers now sitting in homes and collecting dust.

For that reason, Osborne declared, “There will never be an IBM in software.” Because when you’re dealing with $50 software products, the vicariousness of human nature would lead everyone to want to buy something a little different. Kildall agreed, saying that software was where we’d likely see all the real inventors go.

Kildall followed up on Osborne’s remark about smaller inexpensive computers gathering dust. Did Osborne have any idea about, say, the percentage of sub-$1,000 personal computers that were still in productive use? Harp actually replied, saying she believed about 50 percent of such machines were now gathering dust. She admitted there was no way to substantiate that figure as there was little research in this area. But she relied on personal anecdotes from people who had purchased early machines from places like Radio Shack. Once these people brought their cheap computers home, they realized they didn’t quite have the need they thought they did for the system.

Osborne chimed in, claiming his company had done some research in this area and found it tended to vary by brand. He cited the Sinclair ZX80 and ZX81 machines as examples of computers that were mostly gathering dust. Again, Osborne pointed to the price. People purchased those machines for around $300 just to find out what computers were all about. Having fulfilled that purpose, the fact the machines were now collecting dust should not be considered an indictment of the products. Osborne also pointed to the Commodore 64 as perhaps the model where more machines were collecting dust than any other, primarily due to reliability problems. And given the machine only cost $200 in the first place, customers were just not going to pay to have a Commodore 64 fixed when it broke down.

Harp added that her own company, Vector Graphic, had no interest in getting into the low-cost home computer market. She said anything under $3,000 was an absolute no for her.

Cheifet said there had been chatter about a “falling out” starting to take place in the computer industry. He noted many people considered the demise of Osborne Computer an example. Was this in fact the start of a trend? Osborne said that his former company’s failure had nothing to do with competition or an industry “falling out.” He said the company “plain and simple committed suicide,” although he declined to elaborate at this time. As to Cheifet’s question about an industry falling out, Osborne said the answer was partially yes and partially no. IBM, with its 70 percent market share, was clearly dominating the market. On the other hand, if you took that remaining 30 percent, it was huge enough to support a lot of companies that could sell between 3,000 and 10,000 computers per month.

“I Don’t Think Clones Will Be a Long-Term Solution”

Joining the group for the final segment was Gene Amdahl, a legendary former IBM executive who now ran his own startup supercomputer manufacturer, Trilogy Systems. Kildall asked Amdahl for his perspective on the viability of the new IBM Personal Computer clones that were on the market. Could they be successful? Amdahl said he was not that familiar with the PC end of the market, but he was quite familiar with IBM and how it operated.

Amdahl noted the company’s pattern had not changed much over the years. They move into a field once it was proven just well enough in terms of becoming a size sufficiently large to be of interest–and where somebody else has learned what it takes to satisfy the market. At that point, IBM moves in to take its share of the action. Assuming IBM has the same degree of success in personal computers as it had in high-performance servers, Amdahl said the IBM architecture would become a de facto standard–not because IBM had the best architecture, but rather because so many people would expect them to be successful. Consequently, the cost of developing applications for the PC would be sufficiently great to discourage people from changing to another architecture.

Kildall asked how competitors could track a change or transition that IBM might make in the future on a moment’s notice. Amdahl replied you couldn’t do it perfectly, but if you were in a portion of the field where the technology or certain functional limits had formed the boundary of where you could go, as those possibilities moved forward, you could expect IBM’s offerings would likely move in that direction.

Kildall asked Lore Harp for her feelings about the IBM clones. She said they were not going to have a future. Right now, the clones were enjoying a tremendous wave because of IBM’s inability to deliver their machines in sufficient quantities. But most of the clone companies were not creating any value. They were putting hardware together and buying software available to everyone. But they were not adding anything that made them “proprietary.” She believed that if IBM took a different turn, the software companies would follow suit and leave the clone makers scrambling to protect their customer base. And she did not see how these competitors could turn as quickly as IBM, since the company planned its moves a year or more in advance. Harp was emphatic, “I don’t think clones will be a long-term solution.”

Cheifet turned back to Adam Osborne and asked if the computer business was now more driven by advertising and packaging than technology. Osborne said the major difference between the personal computer market today and the high-performance mainframes that Amdahl was talking about was the fact the former really had nothing to do with hi-tech anymore. For that reason he both agreed and disagreed with Harp’s assessment of the market. He agreed the bulk of the clone companies would not make it. But there was still a vast market for clone companies.

Osborne broke it down this way: There was about 15 percent of the market that would wait for Hell to freeze over before buying from IBM. Another 15 percent of the market didn’t care. And the remaining 70 percent preferred IBM. That 15 percent who would never buy from IBM was still a huge market. Osborne added that IBM was shipping 10,000 personal computers a day right now. Even if IBM ended up leaving some of those customers behind, there was room to come in. So if you got a company who understood manufacturing on a low-cost, high-quality basis, they could start eating into IBM’s PC share in a way that nobody ever could on the mainframe side.

Kildall observed this meant the clone companies that would be successful would be the ones that thought of themselves as manufacturing companies rather than computer companies. Osborne said that was absolutely right.

Cheifet brought the discussion and program to a close by asking about the role of venture capital. How was it different now trying to raise funding for a computer company now versus a decade ago? Amdahl noted that during the 1973-1975 recession, when capital gains tax rates were also very high, the stock market for new initial offerings was “abysmal” to say the least and essentially non-functional. Today, there was a rather “buoyant” market. Cheifet asked Harp about her own VC experience. Harp noted she’d been through two rounds of funding, once in 1979 when the capital pool was just starting to grow, and again now in 1984. As far as her company, Vector Graphic, was concerned, they did not have much leverage to determine their price since they badly needed the cash. But in many startup situations, when someone had a good angle, they could essentially name their price because there was not many good deals around, but there was a tremendous amount of venture capital available.

Adam Osborne (1939 - 2003)

Adam Osborne was a chemical engineer by training. Born to British and Polish parents living in Thailand, Osborne later married an American citizen and moved to the United States. He first became known in the tech world as an author and publisher of computer books, including his 1975 text, An Introduction to Microcomputers. From this, Osborne launched his publishing business, Osborne & Associates, which he sold to McGraw-Hill in 1979.

Osborne started his computer company in 1981. Its first and most famous product was the Osborne 1, a 23-pound portable CP/M-based computer with a built-in black-and-white screen. As described in Martin Campbell-Kelly’s obituary of Osborne for the Independent:

A unique selling point of the computer was that it came with a software bundle consisting of a word-processor program, a spreadsheet and other popular application programs; in this way users could have the computer up and running without the need to buy additional software. It was said that the retail value of the software was more than the computer’s $1,795 selling price.

The Osborne 1 predated the launch of the IBM Personal Computer. And while the machine was initially successful, selling about 100,000 units according to Campbell-Kelly, Osborne quickly faced stiff competition from IBM and other early portable computer manufacturers like Kaypro and Compaq.

During the Chronicles episode, Osborne claimed that many people misunderstood the reason for the company’s quick demise. Technology writer Robert X. Cringely elaborated on this in a 2007 article. Essentially, the conventional wisdom was that Adam Osborne destroyed his own company by “pre-announcing” a new MS-DOS machine that would replace the Osborne 1’s existing CP/M model. This purported marketing blunder–alluded to even in the Independent obituary–led to a collapse in demand for the Osborne 1. But Cringely, who spoke with a former Osborne Computer employee, said the real story wasn’t quite so simple:

[I]n fact, [Adam Osborne] told reporters there wouldn’t be a DOS version for at least a year because it was too expensive for his price-sensitive line. He announced the Executive – the follow-on model to his successful first Osborne – one month before delivery, and sales dropped in half for a couple of weeks while inventory cleared out.

Now, demand for both Osborne machines did collapse, but Cringely said the reason for that was the new Osborne Executive “was not competitive with the Kaypro,” which he described as a “slapped-together rival” that cost $200 less than the Osborne 1 and came with a built-in screen that was larger than either Osborne model. Basically, all of Osborne’s customers went to Kaypro. But this was not the result of Adam Osborne jumping the gun to announce an MS-DOS machine.

In any event, Osborne Computer never shook off the failure of the Executive machine. The company entered bankruptcy in 1983. It emerged from reorganization briefly in 1984 and managed to produce a couple of new models. But by 1985 the company was back in bankruptcy and ultimately liquidated

Meanwhile, Osborne’s newest venture at the time of his Chronicles appearance, Paperback Software, also ended up suffering an ignominious demise. For all the talk of IBM clones in this episode, Osborne positioned Paperback as a software clone-maker of sorts. Specifically, he marketed a $99 spreadsheet program called VP-Planner, which was basically a clone of the popular Lotus 1-2-3 software, which cost about $500.

Lotus sued Paperback for copyright infringement. The crux of the lawsuit was that Lotus considered the 1-2-3 user interface–notably the structure and organization of its menus–to be copyrightable material. A federal judge in Massachusetts agreed and issued an opinion in June 1990 holding that Paperback committed copyright infringement. The litigation took nearly three years, during which time Paperback’s sales cratered and never recovered. The court’s decision forced Paperback to withdraw VP-Planner from the market and the company dissolved soon after.

Osborne ended up leaving Paperback shortly before the judge’s ruling. In 1992, he contracted a brain disorder and decided to return to India, where he’d lived as a child during World War II. Osborne died in India in March 2003.

Gene Amdahl (1922 - 2015)

This episode didn’t really touch on Gene Amdahl’s enormous contributions to the computer industry. Jack Schofield, in an obituary of Amdahl for the Guardian, offered some crucial context:

Whenever you draw cash from your bank, take out insurance, book a flight, or shop in a large department store, the transaction was probably handled by an IBM mainframe originally designed by a former South Dakota farm boy, Gene Amdahl, who has died aged 92. Amdahl’s System/360 mainframes, announced in 1964, transformed IBM, and transformed computing. They became such an essential part of large-scale data processing that most Fortune 500 companies are still using them 50 years later.

Amdahl left IBM in 1970 and made multiple efforts to start his own competitor to cut into Big Blue’s share of the mainframe market. His first venture, Amdahl Corporation, developed IBM-compatible machines. Gene Amdahl then left that company in 1979 to start Trilogy Systems. Amdahl Corporation continued on without him until the mid-1990s, when it was acquired by Fujitsu.

As for Trilogy, things didn’t quite go so well. In Part 13 of this series, I discussed a “Random Access” news item about Trilogy delaying the shipment of its promised supercomputers “due to problems with the metal coating of its prototype chips.” That delay proved to be permanent.

Myron Magnet, writing for Newsweek in September 1986, detailed how Amdahl and Trilogy burned through $230 million in capital–including $60 million raised in a public offering–to develop an integrated chip for mainframes:

The project kept missing deadlines, then revised deadlines. Harder to fix, the company wasn’t working quite right either. The parts of the organization never seemed to mesh precisely. Engineers didn’t develop that second-nature reliance on one another that for Amdahl marks a successful technology company. Instead employees galloped off in all directions, painting their own projects as essential and demanding reinforcements. The organization grew bigger and more unwieldy by the month. Like tunnels started from both ends that fail to meet in the middle, the separate projects never cohered into a whole. “We were off only a little,” Amdahl says in his slow, shy manner, “but it was enough.”

In 1985, Trilogy merged into Elxsi, another computer manufacturer. Amdahl remained with the merged company until 1989. He went on to form two other startups, one of which, Commercial Data Servers, apparently still exists today under the name Xbridge Systems, Inc.

“Bored Housewives” Started One of the First Small Computer Companies

Vector Graphic started out in the computer industry around the same time as Apple. Indeed, Benj Edwards noted in a 2015 history of Vector for Fast Company that Lore Harp and her co-founder, Carole Ely, had a booth at the first West Coast Computer Faire in 1977 just a few rows from where Steve Wozniak and Steve Jobs showed off the Apple II.

Edwards described Harp and Ely as “bored housewives” looking to go into business together. After rejecting the idea of starting a travel agency, they decided to start a computer company after Harp’s husband, Bob Harp, had been tinkering with an Altair 8800 computer kit. Bob Harp had developed an 8-kilobyte static memory board for the Altair, which became Lore Harp and Carole Ely’s first product. Edwards said Bob Harp suggested the name “Vector Graphic” because he planned to develop a video board that never came to fruition.

As Lore Harp mentioned on Chronicles, Vector Graphic only had a few thousand dollars in startup capital–$6,000 according to Edwards. At first the company only sold the memory boards. But later, Bob Harp designed several additional components that were then assembled into a full-scale computer:

The result was the Vector 1, launched in 1977, which shipped in two case colors, green or orange (or “rust” as they called it at the time). Lore and Carole’s emphasis on visual aesthetics led them to offer this choice of colors at a time when many companies gave little thought to what their computers looked like (and it was only starting to become a given that PCs shipped in a case at all). An attempt to order orange circuit boards to match the orange case went awry when the first batch of fifty came back pink.

Lore Harp said that although the Vector 1 and Apple II launched at basically the same time, she saw her company as targeting the business rather than the home computer market. “We always looked at the Apple II as more of a toy,” Harp told Edwards. However, when the spreadsheet program VisiCalc launched exclusively on the Apple II in 1979, Edwards said that prompted many business customers to prefer Wozniak’s machine over the Harps’ Vector 1.

Nevertheless, the Vector 1 sold well and Lore Harp was featured in the national media as one of the country’s first prominent female computer executives. Edwards said that Harp “would even end up partying in Paris, France, with California governor Jerry Brown and Steve Jobs as part of a delegation of California technologists.”

But the business took a toll on the Harps’ marriage and they ended up divorcing. Meanwhile, Vector faced an existential threat from IBM’s looming entry into the personal computer market. Lore Harp decided to take Vector public before IBM took away their market share. Bob Harp ended up leaving the company after he failed to persuade his estranged spouse to pivot to manufacturing IBM PC clones instead.

Vector did go public and Lore Harp briefly stepped aside as CEO following her second marriage to Patrick McGovern, the founder of computer magazine publisher IDG, in 1982. Harp returned as CEO, however, in May 1983. By this point, Edwards described Harp’s role as “trying to chart the destination of a sinking ship.” Most of Vector’s business customers had switched to IBM. Shortly after her Chronicles appearance, Harp resigned as CEO for a second time.

Vector belatedly tried to move towards IBM PC compatibility, but the company filed for bankruptcy in 1985. The following year, a holding company liquidated Vector’s remaining assets. Edwards offered a short eulogy: “The truth is that, from 1982 on, no one could have saved Vector. The firm ultimately shared its fate with the every other PC maker that didn’t jump on the IBM clone bandwagon.”

Today, Lore Harp McGovern chairs the McGovern Institute for Brain Research at MIT, which she co-founded with her late second husband, who passed away in 2014.

Woz’s Apple the Lone “Clone Wars” Survivor

Ironically, the only company represented in this episode that survived the IBM clone wars was Steve Woziak’s Apple. Everyone knows that Wozniak designed the Apple I and Apple II computers. He also played an early role in the development of the Macintosh, although that was cut short after Wozniak sustained injuries in a February 1981 plane crash and briefly left Apple. He did return for a stint in the mid-1980s and reportedly remains on the payroll even to this day, but he has not been an active employee since 1985. Wozniak has participated in a number of startup ventures since then, none of which strike me as particularly notable–unless you want to count the Us Festivals.

Wozniak turned 71 on August 11. He was most recently in the news for speaking out in favor of “right to repair” laws, which are strongly opposed by current Apple management.

Notes from the Random Access File