Visit Your Local PBS Station PBS Home PBS Home Programs A-Z TV Schedules Watch Video Donate Shop PBS Search PBS
I, Cringely - The Survival of the Nerdiest with Robert X. Cringely
Search I,Cringely:

The Pulpit
The Pulpit

<< [ Be Absolute for Death ]   |  Bill to Linus: You Owe Me.  |   [ The Other Side of .NET ] >>

Weekly Column

Bill to Linus: You Owe Me.: Did Bill Gates Invent Open Source Software? No, But He'll Take Credit For It, Anyway

Status: [CLOSED]
By Robert X. Cringely
bob@cringely.com

Thanksgiving week is always difficult for me. PBS wants my column early, of course, but the real problem is that much of what I write is often lost to readers, obscured by the effects of whatever that chemical is in turkey meat that makes us fall asleep. People simply don't remember what I write that week, which is of course, this week. I have to work all the harder to shock them out of their holiday stupor. So the shocking questions for this week are 1) Is Bill Gates really the father of the Open Source software movement? and; 2) If Bill isn't the father of Open Source, did he violate a crucial IBM nondisclosure agreement and ought to pay billions in damages to Big Blue as a result?

It is hard to imagine Bill Gates claiming to have started — or even helped to start — the Open Source movement, especially since he has been widely quoted as once describing it as communism. But claiming paternity he kinda, sorta did at Microsoft's recent annual meeting while answering questions from shareholders.

Here is the fateful question: "It appears to me that the open source movement is gaining momentum, and as I understand it, the key to success of a software product involves efficiently building an ecosystem of developers and users, resellers, and so forth. Isn't the open source model a more efficient paradigm for building such a community around your products, and isn't perhaps Microsoft maybe on the wrong side of that trend of long-term?"

And here is the answer from Bill Gates, or at least the first part of it (the complete answer, which is quite long and circuitous, can be found under one of the Links of the Week on this page): "Let me start out, really the reason that you see open source there at all is because we came in and said there should be a platform that's identical with millions and millions of machines, and the BIOS of that should be open to everybody to use, and all the extensibility should be there. And so it was very predictable that once we had gotten the PC going, and going and gotten hundreds of millions of machines out there, that it had always been sort of free software and the universities would flourish and there would be more of that..."

I guess we can forget, then, about MITS giving to the world the S-100 bus, Gary Kildall inventing the ROM-BIOS for his CP/M operating system, and Steve Wozniak creating the Apple II as an open architecture with millions of users. Microsoft even made CP/M cards that could be installed in Apple IIs. And all this was years before the IBM PC and PC-DOS were introduced in 1981.

The gist of Bill's argument is that Open Source requires a large pre-installed base of genetically identical computers, and that base was provided by DOS and Windows. Okay, maybe we can buy that. But then Bill goes on to claim that the BIOS — the Basic Input-Output System that sits between the operating system and the computer hardware — "should be open to everybody to use." Tell that one to IBM, which somehow thought the IBM PC BIOS was their property. They held a copyright on it, after all. Compaq Computer spent over $1 million reverse-engineering the IBM PC BIOS to create the first IBM PC clone. If Microsoft had been working so hard to open up that BIOS, Compaq could have saved their money.

What WAS Microsoft's role in opening-up the PC BIOS? If they were, indeed, pioneers in this effort, then they were, as one of my canny readers suggested, violating an IBM non-disclosure agreement, and would have been subject to billions in penalties. That was long ago, and the statute of limitations has expired, so Bill might well be telling the truth, admitting that he had deliberately undermined his old partner, the company that made Microsoft what it is today. Or Bill could be bending the truth a bit, though I can't imagine why. Frankly, neither answer makes Microsoft look very good.

I needed another source to help me converge on the truth, so I e-mailed Jack Sams in Florida. Jack was the guy from IBM who was sent to Seattle to meet with Microsoft back in 1980, and tell them about the still-secret IBM PC. Jack was also the guy who mistook Gates for an office boy at the start of that meeting.

And Jack had a lot to say on this subject:

"Bill responded like a true politician by switching the question from Open Source to Freeware (ugh), then to Open Architecture (read de facto standards), which he claims to have prompted IBM to adopt for the PC," said Sams. "Bill did, in fact, influence the IBM PC interface architecture, but our "open architecture" decision was ab initio.

"Your (reader) challenged Gates' claim by noting that the (IBM BIOS CHIP) is highly proprietary. He didn't distinguish the copyrighted chip from the interface architecture it implements. The chip is indeed copyrighted and could be infringed. The open architecture it supports was extended by Paul Allen's DOS 2.1 to actually support dynamic addition of features and capabilities at run time. This (DOS +BIOS) open architecture has been public domain since it first shipped ( Byzantine, but open).

"So, everyone is more or less right. Bill remains an artful dodger and a selective rememberer, but, aren't we all?

"Here's the open architecture/BIOS history as I saw it during 1980.

"IBM's (August 1980) product development plan for the PC assumed almost from day one that we would have to rely on a number of independent, third party hardware and software developers to respond to the demands of a mass market. (think programming languages, word processors, games, spreadsheets, joysticks and classroom drill, a million machines and a three year program)... We consciously intended to host other vendors independently developed software, and we were almost completely dependent on third party peripheral devices (color video, disk, tape, printer, communications) because our own available I/O was hardwired for EBCDIC date encoding.

"The product strategy demanded a reliably defined interface that allowed other vendors' hardware to physically and logically attach to the PC bus and for other vendors' software to access all system services. The assembler source code that implemented this "BIOS" was written by David Bradley, of IBM, in Seattle, in consultation with Microsoft. Its design was limited by three givens:

  1. Microsoft's 8086 version of ROM BASIC.
  2. The 8 bit I/O bus and device control architecture inherited from the IBM Datamaster (a failed earlier attempt to build an IBM PC).
  3. The existing 8088 motherboard design that was planned as an upgrade for the Datamaster.

"These hardware "bootstraps" were never acknowledged in the "official histories" by IBM, and may well have been concealed from Bill Gates until early 1981... hence his claim to have persuaded us to use a 16-bit chip. Somebody nodded wisely and said "good idea", but the 8088 prototypes were already running in August with IBM 8-bit I/O.

"The BIOS code was written very early (in September/October 1980), during the first (consulting) agreement we negotiated with Microsoft in August 1980. It established the infamous 640K memory boundary and other simplifying conventions to allow the system to be run entirely from ROM. IBM copyrighted the CHIP and published the interfaces at first customer shipment. (I'm sure there were arguments against publishing until the last minute, and Bill would certainly have had an input; but that's just my opinion, I was out of the loop after November 1980)

"At this level, Bill Gates can certainly claim to have "influenced" the open architecture strategy. He was our consultant, he had practical experience interfacing BASIC to a succession of systems with a variety of ASCII I/O devices and device controllers, and he was the first, we expected, of many vendors whose products would become replaceable parts in PC systems.

"However, the "open architecture" strategy was entirely deliberate on IBM's part. We expected to defend our own hardware market:

1) By being the lowest cost producer of the core system, and
2) By asserting copyright protection for the bios chip(s).
3) By quickly offering a series of cheaper, faster, better upward compatible systems and upgrades.
4) By staying out of the PC software development business.

"All were relatively successful strategies through the PC, PC/XT, and PC/AT, although our reliance on overseas suppliers set the stage for the PC/AT clone takeover as soon as there was a reliable source of reverse engineered BIOS chips.

"When the PC Division began to plan an 80386/AT in 1983, the IBM Corporate Management Committee took the business back from Don Estridge and directed its new management to redevelop the PC as a proprietary IBM product with "normal" profit margins and a full range of proprietary software and I/O.

"So sad."

Bill dodges another one.

Those interested in slightly more recent computer history might want to know about next week's celebration of the first 10 years of QuickTime, Apple's extensible multimedia technology. The amazing thing about Quicktime is that there was nothing like it before, and everything has been like it since. Look at the guts of Real Player or Windows Media Player, and you'll see structural copies of QuickTime. It is very hard to be an original, to be the first, and to still survive a decade later, but Quicktime does all that. And it might even get the last laugh. Apple is rumored to be preparing an MPEG-4 player for Quicktime (the Quicktime file format is already used by MPEG-4), which ought to give the system perpetual legs and a real advantage against more proprietary solutions from Real and Microsoft.

Back in 1990 when Apple first conceived of QuickTime, the world of "multimedia" was one of laserdiscs. A multimedia application was a Hypercard stack connected bya serial cable to a Macintosh. The stack let you navigate to a particular clip. The video was then played on a TV screen. In 1991, the big step forward was to display that video on the computer screen... but you still needed the laserdisc. And when Apple management announced QuickTime in 1990, the idea was very much about perpetuating and supporting that model. The QuickTime team subverted that whole system by saying that every PC (Mac!) should be able to play video on its own... no special hardware. So they developed software only video and audio that scaled itself to match the users machine. Constrast this to the PC industry which in 1992 was obsessed with multimedia PC, which was basically just a sound card plugged into a stock PC. Miles apart. The QuickTime work put in the foundation for ground breaking titles like MYST and Peter Gabriel's XPLORA 1, and set the model that was later followed for MPEG-4 and DVD.

Work on QuickTime 1.0 was completed on December 2, 1991, with a product launch at the San Francisco MacWorld show in January 1992. There will be a party, of course, held on December 1st in San Francisco to celebrate the anniversary. Proceeds are earmarked to raise funds for a permanent QuickTime museum exhibit. And all this is because a nonprofit, all-volunteer organization called Friends of Time is determined to secure the place of QuickTime in technology history. Their web site is listed among the Links of the Week.

Now if only there was a similar group called Friends of Bob.

Comments from the Tribe

Status: [CLOSED] read all comments (0)