PC pioneer Gary Kildall's unpublished memoir revealed

CP/M and BIOS inventor adds to story of how Microsoft's IBM deal shaped the PC forever

The Computer History Museum has revealed part of an unpublished memoir by Gary Kildall, a programmer and entrepreneur who made critical contributions to the personal computer industry in its formative years.

Kildall died aged just 52, in 1994, but in his short life he earned fame for developing CP/M, an operating system that became an early de facto standard for PC operating systems. As more people wanted to run CP/M on more systems, Kildall promoted the idea of a BIOS that would let computers boot into a simple state from which a full operating system would be loaded.

Kildall's also remembered for probably giving Microsoft its big break, by not taking immediate interest in IBM's offer to licence CP/M as the OS for the product that would become the IBM PC. Kildall's hesitation led Big Blue to ask a couple of kids called Gates and Allen – Bill and Paul – if they had a comparable product. Microsoft never looked back.

The account above of Kildall's encounter with IBM is brief, not least because there are conflicting versions of the story, and this memoir is not the place to sort through them. Industry historians are doubtless slavering at the prospect of an as-yet unpublished book shedding light on those events.

Sadly, this portion of the memoir touches only lightly on those events, in passages touching on the “Microsoft Softcard”, an add-in card that Microsoft developed to bring its BASIC, and CP/M, to the then-hit Apple IIe.

Kildall says Gates tried to buy CP/M outright, but he refused and instead sold 10,000 licences at $2.50 apiece.

“Bill's Microsoft had a BASIC interpreter that would operate with the Apple computer using his Softcard and my CP/M,” Kildall wrote, but says Microsoft wasn't happy.

“To Bill's dismay, the Apple platform for application software was controlled by Apple Computers, not Microsoft. That's a wild card for Bill. He 'needed' to control the platform for his BASIC, and IBM was to provide him with that.”

In a a section titled “The Early Days of Microsoft” Kildall said he didn't trust Gates.

“I was always apprehensive of his business moves,” Kildall wrote, “as I found his manner too abrasive and deterministic, although he mostly carried a smile through a discussion of any sort.”

But the manuscript suggests that after the Softcard deal, the relationship soured, as described in this passage:

The combination of Kildall and Gates could have been a killer-deal in those days. I had the operating systems for the decade to come, and he had the opportunistic approach to garner business. But, our attitudes differed entirely, and that could also have been a disaster. I think we both realized this and simply let the 'deal' die.

The excerpt does dispel what Kildall calls “an old industry rumor” that he “came to Intel and offered CP/M for their microcomputer operating system, and Intel rejected it."

Kildall's version of events has Intel deciding that all OS development should be done in-house, so he didn't even get a look-in. He says that was a blessing in disguise, because elsewhere in the manuscript he criticises Intel for inflating products and deliberately closing ecosystems. CP/M, he explains, succeeded because it was cheap and could run on a variety of hardware.

Elsewhere in the 78-page-extract from Computer Connections: People, Places, and Events in the Evolution of the Personal Computer Industry, Kildall recounts his education and early work in the industry and describes a gentler time in the industry.

For example, he describes how he had a hand in a critical decision that helped Intel make the 8008 processor a hit.

One day, I knocked at the door of Hank Smith's office at Intel. He invited me in. Hank was the manager of our minuscule microprocessor software group. I told him that I could make a compiler for the 8008 so that his customers didn't need to use low-level assembly language. Hank didn't know what that meant, but I showed him how a customer could write:

X=Y+Z

and that would make several lines of assembly language. He immediately got on the phone and called a customer he was courting. The customer liked it, and Hank, with a smile, said to 'go for it.' I like corporate decisions like that.

Another quaint passage explains an office party during which Kildall went to fetch more booze while wearing roller skates. “I found temporary haven by hugging a telephone pole near the liquor store for refuge” is one of the extract's more memorable sentences.

Elsewhere in the extract he points out that he should really be pointing out which product names he uses are trademarked, but then remarks he's not using his own trademark on CP/M. Your correspondent suggests that quip, and much more of the memoir, would have been significantly tightened by an editor.

The Museum says Kildall penned the memoir in 1993 and circulated it among friends at Christmas, intending it to be published in 1994.

His untimely death meant the book was never completed.

The Museum says the 78 pages today is just one portion of the work and that it plans to release more.

If you'd like to read the extract, a big red button at the bottom of this page leads you to a form on which the Museum asks that you promise not to distribute or reproduce it. ®


Biting the hand that feeds IT © 1998–2017