Inside the EU Court's Ruling: Microsoft's Abuse of Dominance

This morning's ruling from the European Court of First Instance dissects both Microsoft's and the European Commission's arguments with logical, plain-spoken precision. It makes the point that Microsoft doesn't get to decide what "interoperability," "bundling," and "evolution" mean. But neither does the EC.

As an institution unto itself, the European Union is frankly not much older than Microsoft. America has a few hundred years of legal precedent to guide it, and the EU characterizes its heritage as the combination of thousands of years of legal foundation among its 27 member countries. But in actuality, the European Commission is bound by its own principles not to apply itself to any case, even if it affects the whole of the continent, unless it can prove to its people that it can do a better job than the legislative authorities of its member states can do collectively.

So unlike America, Europe has both the freedom of being able to set new legal precedent for itself, and the burden of having to do so when necessary. In the case of Microsoft, the same test of whether Microsoft's conduct should even fall under special legislative oversight or jurisdiction established by American antitrust law - which deals with specifically defined monopolies - doesn't apply to European law. There, the test of whether a company's conduct falls under special legal authority rests on whether it dominates the market.

Some believe the European test to be much more realistic than the American one. If you consider it thoroughly, it doesn't make much sense on the surface for a monopoly to have a negative impact on its competitors, since by the basic definition, a monopoly has no competitors.

But the original US antitrust trial against Microsoft was about "barriers to entry" - how the company maintained its monopoly status by excluding others from participating. The EU trial is about something very different: "abuse of dominance," which is far broader and not at all based on American precedent. The difference between a monopoly and a dominant power is as broad as the difference between "overwhelming evidence" and "preponderance of evidence" - between upsetting the scales and just tipping them slightly.

Another issue at hand is how the EC makes new laws where little or no precedent previously existed. Specifically, can the EC make a law defining what a "server" is or should do? The Commission found in 2004 that Microsoft should have shared its interoperability protocols with competitors because, by definition, a product that purports to be a server should be in the business of sharing.

From the EC's perspective, if you claim to make a server and it's not sharing, that claim may be fraudulent. From Microsoft's, what right does any legislative assembly have to tell Microsoft what a server should or should not do?

But there's also another huge issue on the table, one which looms over the entire proceeding but which officially, at least for now, is being treated as though it were not in doubt: the extent to which a European legislative body can impose restrictions upon American companies. Any company that does business with European customers must abide by European law; the same holds true for the US and most other countries.

Yet the original complaint in this case with regard to server interoperability was filed by Sun Microsystems, over then-alleged conduct on the part of Microsoft that certainly affected Sun's products in Europe, but in the rest of the world as well. Can an American company taking part in a worldwide market seek redress for its grievances in whatever court upholds the strictest jurisprudence?

Point 1: Microsoft didn't share

Sun originally complained that it was not receiving enough information from Microsoft to be able to develop a server operating system that interoperated with Windows. The European Commission found that complaint to be valid, and in so doing, rendered a decision whose purposes included legally defining Windows: "The proper functioning of a Windows work group network relies on an architecture of client/server and server/server interconnections and interactions, which ensures a transparent access to the core work group server services," reads the EC's 2004 decision.

Microsoft's counter-argument was that all the information Sun needed to build Solaris to be interoperable with Windows was already available online through MSDN. That level of information constituted "interfacing" with Windows. But the level of "full interoperability" the EC wanted Microsoft to provide, the company argued, would have amounted to it giving away its trade secrets - everything a competitor might need to create its own Windows.

In other words, if a competitor gets to know not only what a server communicates but how, then that competitor wouldn't need to interoperate with that server - it could go out and build its own, using the "how" it just learned.

The Court of First Instance didn't buy that argument at all. "Microsoft claims that the Commission's concept of interoperability means, on the other hand, that its competitors' operating systems must function in every respect as a Windows server operating system," reads the Court ruling this morning. "That situation could be achieved only if those competitors were allowed to 'clone' its products, or some of their features, and if information on the internal mechanisms of its products were communicated to those competitors."

In short, the "what" doesn't imply the "how."

Next: Point 2: Interoperability is a Standard