Last Updated on July 4, 2018 by Dave Farquhar
The urban legend says Gary Kildall snubbed the IBM suits by making them wait in his living room for hours while he flew around in his airplane, and the suits, not taking it well, decided to cut him out of the deal and opted to do business with Bill Gates and Microsoft, thus ending Digital Research’s short reign as the biggest manufacturer of software for small computers.
Gates doesn’t talk about the incident. About all anyone ever managed to get out of him was, “Gary went flying.” That’s the likely source of the urban legend. I’ve never found Kildall’s account of the story, and he died in 1994, so he’s not talking.
Symantec founder Gordon Eubanks was around at the time and knew both Gates and Kildall, and his recollection of what happened is pretty different from the story most people have heard.
Eubanks explains it pretty simply. Kildall didn’t handle the business deals. Period. His wife did. Kildall was a free spirit who wanted to write software on his own terms. He didn’t have much business sense and didn’t want it. When IBM came knocking at the door, his wife was talking already to Hewlett-Packard, who was Digital Research’s biggest customer at the time.
Hindsight says you tell HP to get lost and talk with the 800-pound gorilla. But remember, in 1980, IBM expected to sell a total of 50,000 IBM PCs. That’s it. Even IBM didn’t expect to become Kildall’s biggest customer. (The Z-80/8080 version of CP/M had, at the time, sold about 600,000 copies.) Since IBM was going to use an 8088 CPU, the OS would have to be reworked, so IBM’s 50,000 copies wouldn’t exactly be an easy 50,000 copies to produce. So it’s not difficult to see why Kildall was less than ready to drop everything for IBM.
By other accounts, neither Kildall nor his wife nor their attorney wanted to sign the non-disclosure agreement IBM had all its other partners in the project sign. Signing NDAs probably went against Kildall’s nature; his wife and their attorney feared that it was so restrictive it would be easy to unintentionally violate it and open the company up to legal action it couldn’t afford.
The situation was considerably different at Microsoft. Microsoft’s dominant product at the time was Microsoft Basic, as in the computer language. Its Basic was available for virtually every computer and was pretty much the industry standard. Microsoft was a smaller company with a smaller number of products and
The situation is also much better documented. It gets considerable play in Stephen Manes’ 1993 book Gates: How Microsoft’s Mogul Reinvented an Industry-and Made Himself the Richest Man in America. By Manes’ account, Gates was willing to do anything IBM asked. Gates, unlike Kildall, took the IBM PC seriously. In Kildall’s defense, IBM had consulted Gates on the design and Gates actually had access to a prototype. IBM had tried to enter the microcomputer field before and aborted those attempts, so Gates had a better picture. And Gates’ cooperation in the design of the IBM PC would leave Microsoft in a good position if it had only sold 50,000 units, or less. IBM intended to do a successor. Gates had pushed for, among other things, a 32-bit Motorola 68000 CPU in the IBM PC. The 68000 was far and away the most advanced CPU on the market at the time, and it could run Microsoft’s top-flight operating system, Xenix. Yes, if Bill Gates had gotten his way in 1981, Microsoft probably would be the biggest Unix vendor today. Instead, Xenix ended up with a tangled history of its own, including an indirect connection to Kildall.
If the IBM PC had flopped, Gates could just say “I told you so,” and start pushing again for a PC design utilitizing the 68000 and Xenix.
Microsoft had nothing to lose. Digital Research had everything to lose.
IBM had its mind made up that the IBM PC would use the Intel 8088 processor–a cheaper version of the 8086 that had an 8-bit data bus–but they needed an operating system and they didn’t know if they could get it from Digital Research. The 8088 wasn’t powerful enough to run Xenix.
Microsoft, of course, ended up providing IBM with an operating system. A friend of a Microsoft employee had designed an 8086-based computer for Seattle Computer Products, and quickly wrote an operating system, known as QDOS or SCP-DOS, for the machine. Microsoft bought QDOS for $50,000 and licensed it to IBM. Kildall called it a ripoff and insinuated that it contained some CP/M code. There are numerous stories about this. In the Bob Cringely book Accidental Empires, Kildall is quoted as saying, “Ask Bill why function code 6 ends in a dollar sign. No one in the world knows that but me.”
And in 1996, John C. Dvorak revealed another mystery: He mentioned that he knew someone who had a very early copy of MS-DOS that contained an easter egg that printed Gary Kildall’s name. CP/M had the same easter egg. I’ve looked, but I’ve never heard anything else about that story.
The IBM PC, of course, changed the industry. Illegal or not, MS-DOS went along for the ride and became the foundation of Microsoft’s empire.
Digital Research was bought out by Novell in 1991, then was resold to Caldera five years later. Several spinoffs and buyouts later, the direct descendant of the product that Kildall and Digital Research eventually did end up writing for IBM is now owned by DeviceLogics. And the name Digital Research has been recycled by a reseller of CD-ROM drives and other peripherals in consumer electronics stores.
How the mighty have fallen.
2 thoughts on “How DOS came to be IBM’s choice of operating system”
Wow! Thanks for this, Dave. I remember GEM too, and I always preferred it… and never understood why everyone was using Win 3.1. Meanwhile, I ran GEM and DOS – because Win 3.1 and GEM just weren’t enough for me to get away from the command line.
Command lines make life so much easier. Well, for me anyway. 🙂
Thanks MUCH for finally doing a definitive debunk of this popular Urban Legend. I have long suspected it to be too good to be true, but have never had the authoritative, documented version of the story.
Comments are closed.