June 22, 2026

Over the years, there have been an untold number of books written about IBM and its computing legacy. (Our founding editor, TPM, has most of them in his library.) But very few, if any, of them focus much time on the line of midrange servers that came out of Rochester, Minnesota, including the AS/400, which turned 38 years young yesterday. That void is what drove former IBMer Bill Shaffer, who spent his entire career in the midrange, to write the book IBM & Computing.

Shaffer spent 40 years at IBM, beginning as a programmer working out of the Miami, Florida office and culminating in the IBM printing division out of Boulder, Colorado. In between, he spent many years living and working in Rochester, as well as from the IBM facility in Atlanta, Georgia.

When Shaffer retired in 2015, he began writing a book about one of his favorite topics: cars. Like many Americans, Shaffer and his family owned a variety of automobiles over the years, and Shaffer – who caught the writing bug as a contributor to the midrange trade press – decided to document it.

Shaffer’s first book, Shifting Gears: One Family’s Journey Through the Automobile Age, was published in 2021. As he was writing that book, he read several histories of IBM. One particular book published in 2019, James Cortada’s IBM: The Rise and Fall and Reinvention of a Global Icon, caught his attention.

“I read the book, 750 or 800 pages, and there was like one paragraph on midrange, which is where I spent my entire career,” Shaffer told IT Jungle. “I called him and chatted with him and I said, ‘How is this possible?’”

Like Shaffer, Cortana also had a 40-year career with Big Blue, but he spent the bulk of his working with the mainframe and professional services. Shaffer said Cortana was very polite, but defended his view.

“He said, ‘Well, if you want that story told, you have to tell it yourself,” Shaffer said. “And so that was really the impetus to move forward.”

A Thorough Big Blue History

The result is IBM & Computing: The Story of the Iconic Company Amid Technological Progress from the Abacus to AI. Published by The Widsten Press in 2025, the 852-page book offers a thorough review of the history of computing, starting with the pre-IBM creations of Charles Babbage and Blaise Pascal, up to the latest IBM System Z and Power Systems machines, as well as IBM’s emerging quantum systems.

Shaffer offers an insider’s view into the cutthroat business of computer engineering and its intersection with the demands of the market, whether its commercial, scientific, or military. He describes some of the early IBM machines from 1930s, like the IBM 601 Multiplying Punch released in 1931, the IBM 285 Numeric Printing Tabulator released in 1933, and the IBM 402 plugboard released in 1934.

Shaffer discusses the impact IBM machines had on World War II, such as the IBM Automatic Sequence Controlled Calculator, or the Mark I, which was used for calculating ballistic tables by the U.S. Navy, as well as the development of the atomic bomb in the Manhattan Project at the Los Alamos lab.

While this is an IBM-focused book, you can’t really tell the story of IBM without mentioning competing systems. He touches on some famous non-IBM machines, including Colossus, a system created by a team in the UK to break the Nazi’s Enigma encryption machine, dubbed project Ultra. We learn that Shaffer’s father was an American liaison officer working on the Ultra project, which remained a secret until former British intelligence officer FW Winterbotham wrote about what is now understood to be the world’s first digital computer – predating ENIAC by several years – in his 1974 book, The Ultra Secret.

Clearly, Shaffer’s book is not entirely about the IBM line of midrange servers. A good chunk of the meticulously researched book (he offers nearly 50 pages of references at the end of the book) documents IBM’s post-war development under Thomas Watson Jr. and the machines that came out during that time, such as the traditional punch card machines like the Card Programmed Calculator (CPC) systems, the development of newer stored-program computers, like the IBM 650 Magnetic Drum Calculator (MDC), as well as weird hybrids, like the “Wooden Wheel,” or the X795, which combined vacuum tubes and a plugboard.

As computing moved past electromechanical stage, Shaffer tells his readers about the hugely successful IBM systems of the 1950s, such as the legendary IBM 1401. We hear about new programming languages, like COBOL and PL/1, and encoding mechanisms, like Extended Binary Coded Decimal Interchange Code (EBCDIC), which was based on old Hollerith codes.

Future Midrange Glimpse

We start to get glimpses of the IBM’s midrange future in the 1960s. Shaffer documents the huge $5 billion bet (equivalent to $40 billion today) that IBM made in the 1960s to develop the System/360, as part of its New Product Line (NPL) effort. The System/360, which debuted in 1964, went on to be a massive success, driving an estimated $100 billion in revenue for IBM, cementing its reputation as the preeminent system developer, and creating a legacy that continues to this day with the System Z mainframe.

Despite the financial success, it was not all rainbows and fairy tales in 1960s for IBM, Shaffer said. “It was a dark time,” Shaffer said in an interview. “There was a continuing theme or a problem with IBM, which is that they had a proliferation of different computers instead of just one computer.”

While the System/360 was a giant success, it neither scaled high enough nor low enough to meet all requirements, which led to the proliferation of incompatible platforms. IBM sought to fix this dilemma with its Future System project, which the company launched in 1971. The Future System included IBM engineers from across the company, including a young electrical engineer named Frank Soltis. The computer would be revolutionary in its design, featuring single-level storage, a 48-bit memory space, and an integrated database.

By 1974, the Future Systems project was in dire straits, faced with scheduling delays and cost overruns. IBM killed the project in February 1975. The prior month, IBM announced the 16-bit System/32, which Shaffer documented in a chapter titled “Glen Henry’s Entry System.” While not a Future Systems project, the computer built at the IBM plant in Rochester, Minnesota would find great commercial success as a follow-on to the System/3, which debuted in 1969 and also was a commercial success. In 1977, IBM launched a multi-user follow-on to the System/32 dubbed the System/34, which introduced small integrated disk drives, twinax cabling, and the 5250 display (to replace the 3270 displays they were using).

While the Future Systems project was dead as of 1975, some of the concepts would be resurrected. Fresh out of college with a PhD in 1969, IBM tasked Soltis with the job of revamping the System/3x line. Soltis and his team of engineers, including Dick Bains and Roy Hoffman and sometimes Henry (when he wasn’t occupied with the System/32), worked for years developing a plan. This R&D work took place in Rochester, but drew the ire of other parts of IBM.

“IBM Research did not contribute at all,” Shaffer writes in his book. “The mainframe folks were borderline antagonistic. Henry describes them as ‘the enemy’ as they were constantly talking about driving the S/370 line down and encroaching on Rochester territory. Though supported by CEO Frank Cary, it was clear that Rochester was under the microscope in running such a large and risky program.”

The Future Systems concepts (if not the project itself) manifested in Rochester’s next creation, the System/38, which launched in 1978. Led by Henry with Soltis as the lead engineer, the System/38 targeted larger companies and was a huge success. Following the launch, Henry moved to Austin, Texas, “in part to avoid the long and brutal Rochester winters.”

Fort Knox And Silverlake

However, by 1982, IBM had a familiar problem: Too many computers. In addition to the System/3x line, IBM had the Model 4300 and the Model 8100, both of which were based on the System/370 architecture, as well as the Series/1, a 16-bit minicomputer introduced in 1976. IBM began a new effort to consolidate the machines, dubbed Fort Knox. Meanwhile, IBM launched the System/36 in 1983 as a follow-on to the System/32 and System/34 entry-level machines.

“The concept of Fort Knox was to combine all the midrange boxes together,” Shaffer said in the interview. “And again, it wasn’t possible. So some of the Rochester engineers that were part of the Fort Knox plan came back and said, you know, we should do a mini-Fort Knox. We should combine the ‘34, ’36, and ‘38 into one system, which became Silverlake, or the AS/400. And that used the Future Systems architecture, sort of like the FS architecture in the flesh.”

Shaffer provides a thoroughly documented account of the Silverlake project and the development of the AS/400, which IBM famously launched on June 21, 1988. He gives the readers a glimpse of the internal politics and personalities within IBM, as well as its desire to compete with DEC and other minicomputer makers.

He reminds the reader that the AS/400 was one of the most successful computers ever created. While the IBM i business is a shadow of its former self, the IBM midrange server was at one time a real market monster. At its peak, Rochester employed 8,000 people and helped drive $14 billion in revenue per year for IBM. If the AS/400 business was removed from IBM at its peak in 1991, Shaffer reminds the reader, it would have been the second largest computer company in the world, trailing only the rest of IBM.

While it occupies only a few hundred pages of the considerable 852 pages of the book, the story of the AS/400 is the real tale that Shaffer sought to tell.

“Obviously, I wrote the book with one particular target market in focus, and that was people who had worked in this area, and maybe had read [Cortana’s] book and came and came away with the same reaction that I did,” he told us.

“As you know, we had this amazing part of IBM that did so much. And it just kind of evaporated in Rochester. Rochester had 25 or 30 buildings at one time. They’re probably down to one building and just a small staff. It’s almost heartbreaking how things have turned out,” he continued. “I wrote the book to provide a more balanced, complete history of IBM and just understand where did IBM come from and what were the problems and what was this thing called midrange and tell the whole story.”

To read more about Shaffer’s book or to purchase it, go to williamcshaffer.com.

RELATED STORIES

Dr. Frank Reminisces on 35th Anniversary, Looks to Platform’s Future

Talking System Architecture With The Frank Soltis

A Frank Solstice

The Long Play

You’re Only As Old As The Applications You Feel

A Platform Of A Certain Age And Respectability

Big Blue Gives IBM i Shops A Special 30th Birthday Bash Box

30 Years And Just Getting Started: IBM i Celebration Looks Ahead

Seven Bright Spots To Ponder On The AS/400’s 29th Birthday

Power Systems GM Weights In On AS/400 Birthday

The AS/400 At 28: A HENRY, Not A DINK

The AS/400 Turns 27, And Still Has Much To Teach IT

Reader Feedback On The AS/400 Turns 27, And Still Has Much To Teach IT

Still A Community Of Common Interest

TFH Flashback: A Community of Common Interest