Lest you think this is going to be another rubber stamp condemnation of Microsoft, I intend to offer a balanced perspective based on the history of the industry, as well as from inside the trenches today. Thus, most of the inflamed rhetoric you have doubtless encountered in the popular media is going to be disregarded because I feel that it is important that each person who has a stake in the industry have a clear picture of what the issues are, untainted by politics, prejudice, and media hype. It also happens that I believe every man, woman and child has a stake in this industry, for it is changing the way every life on this planet is lived. To keep the subject manageable, I will be treating only the PC environment in this discussion. Servers and Mainframes need not apply.

Let’s consider, first, the technical definition of a monopoly. Simply put, a monopoly is when you have no other options. You must deal with a single company if you want a certain type of product. That is why a cartel is not technically a monopoly: The participants in a cartel are all individual companies that have privately agreed on a common pricing or service level. While cartels are illegal in the United States, they are also very difficult to prove in court. A practical contrast would be the previous "Ma" Bell Company, which seems to be gathering herself back together, and the petroleum industry. Ma Bell was a monopoly: there was no alternative if you wanted phone service. The petroleum industry is a different story. There are many different companies, buyers and suppliers. But prices at the pump frequently remain artificially inflated because the companies can privately agree to keep their prices high. This is cartel-like in it practice, but extremely difficult to legally prove in court. Since a monopoly is the sole source in a certain industry, the most frequent side effects are skyrocketing prices and flagging innovation, for without any competition the monopoly has a captive audience. The only feature that improves is reliability. If a company has a monopoly on a product or service, it can dictate how the product or service will be offered. This commonality implicitly breeds efficiency of existing processes, though it certainly doesn’t encourage innovation of new ones. Since Ma Bell handled the whole phone system for the entire country, there was a striking and unimaginative sameness about the entire infrastructure that allowed any technician from any region to go to a different part of the country to work and find the systems and infrastructure to be virtually identical.

The reliability factor of a company is important to consumers. When the Federal Government broke up of the Bell Company, my mother’s tirades blistered paint in the next county. It took a month for my eyebrows to grow back. She had good reasons to holler: the unregulated free market long distance industry was a bramble of technical problems for the first two years after the breakup, and still causes service nightmares for consumers. Reliability and standards were sacrificed on the altar of lower prices and "more choice." But when none of the "choices" work as advertised, what has the public gained?

While I am not an advocate of monopolies, I can certainly see some advantages to them. I work in a computing industry battered by competing standards and systems, constrained by cartel-ism, and crippled in its progress by the practice of turning out inferior products for the sole purpose of beating the competition to market. This makes choices for consumers to be very painful when the standards change more quickly than they can afford to replace their systems. It can be even more expensive to change from one system to another when a poor decision was made earlier.

When Microsoft entered the big time computing scene in the late 1970’s, IBM was under investigation by the Department of Justice for (guess what!) monopolistic practices. Microsoft virtually gave away their license for MS-DOS to IBM in 1981 as "PC-DOS" and the rest is history. Bill Gates wanted MS-DOS to become the de facto Operating System for the IBM built Personal Computers, rather than make a wad of money licensing it for the short haul. This practice has been a long held formula for getting your products into the mainstream by making them the path of least resistance or expense. You know, free stuff. It worked for Odysseus: Offer a wooden horse to the Trojans and you get to sack their city.

In 1982, IBM’s first personal computer was just called the "IBM PC" and boasted a 16 bit Intel Central Processing Unit. AutoCAD v1.0 was announced six months later at COMDEX in Las Vegas in November, 1982. [See Sidebar]. While the Department of Justice investigation of IBM for monopolistic practices was eventually dropped, it was dropped because the torch had been passed. IBM’s monopolistic hold on the market was essentially broken by their inability to keep up with market changes. Microsoft and Intel became a de facto cartel by exploiting the open architecture that IBM established with the IBM PC to the point that other PC competitors (like Hewlett-Packard, Xerox, DEC and Texas Instruments) couldn’t build a machine without using Intel CPUs and a Microsoft Operating System. While some knockoff DOS systems were offered (like Digital Research’s DR-DOS) they had compatibility problems sufficient to keep them out of the mainstream market.

During this era, any company who wanted to produce a machine could walk right into the market and set up shop. IBM was the leader, but there was room for more than just Big Blue in the manufacturing of PCs because they were all based on open standards that IBM had popularized and distributed with a very reasonable fee to license its patent portfolio. Open architecture meant that the creation of new peripherals based on a common expansion interface could be made and marketed by anyone. The PC was fast becoming a fixture of an industry that had previously been for the pocket-protector elite. True, personal computing: A Golden Age, right?

Well, yeah. Briefly.

Since only Intel made the CPUs, it charged whatever it wanted. Microsoft only offered MS-DOS for Intel based hardware. Consequently, application developers wrote to this dual standard out of necessity. The only alternative was to abandon the IBM architecture entirely and trade off for the proprietary extremism of Apple’s Apple II or Macintosh which represented only a fraction of its market potential. The Microsoft-Intel alliance was great because, like Ma Bell, it standardized the industry. It leveled the computing playing field in all markets except Operating Systems and CPUs. It’s the "except" that caused problems.

With the PC market going through its first big explosion in 1984, there was a lot of money to be made and Microsoft-Intel owned the OS/CPU market for the Personal Computer. IBM wasn’t happy that many of the IBM PC compatibles made by other companies were stealing a larger market share than was predicted. So it made what many analysts think was its most fatal blunder ever: It tried to put the genie that was PC open architecture back in the bottle.

Believing that closing the architecture was the only way to get the market back under its thumb, IBM sought to change the physical architecture away from the open standard, replacing it with a proprietary system, thereby freezing out the competition and rendering everyone else’s "IBM Compatible" incompatible. All expansion slots, peripheral connectors and interfaces were to be proprietized. This new architecture would put all the other manufacturers behind IBM while they retooled, and IBM would recover its sales lead again. This new wonder-box would even have its own proprietary Operating System that wouldn’t be reverse compatible to the older hardware. It would be known as OS/2 and was to be co-developed by IBM and Microsoft for the new hardware, called PS/2.

The strategy failed miserably.

IBM was still thinking like a mainframe company. It made design choices for both hardware and software based on mainframe-centric thinking. This yielded systems that were unacceptably slow or unwieldy in the fast moving PC market. When IBM introduced its Micro Channel Architecture (MCA) expansion port on the PS/2, the industry yawned because it had developed its own enhanced port in IBM’s absence. Additionally, Microsoft got so sick of trying to write what was essentially a mainframe Operating System for the PC that it opted out of the co-development process and went back to focusing on Windows which it had initially begun work on in 1983. Since Windows was built for the PC from the beginning, it prospered while OS/2 foundered.

So, at the beginning of the 90’s we have the founder of the architecture (IBM) out in the cold, Microsoft riding high with Windows 3.0 and its bevy of productivity applications (which are slowly squeezing out competitors like Ami Pro, WordPerfect, and Lotus 1-2-3), and Intel cranking out 80386 CPUs for machines as fast as new manufacturing leader Compaq can build them.

But just around the corner, a little semiconductor company waited in the wings to break the cartel and unleash the complimentary specters of monopoly retrenchment and rabid competition to give us the worst of all possible worlds. It all hinged on the interpretation of a key phrase in a license contract with Intel. For the details on how this happened, and how the fallout from that event could result in a happy ending for all of us, be sure not to miss next month’s article!

Peace,

WebWalker