Insights hed: dek: by James Mathewson

A couple of months ago I did two stupid things in an online column. I made a bold prediction that Microsoft would get skewered in court and I promised I wouldn't talk about the Microsoft antitrust case anymore. After putting that column up on our site, Senior Managing Editor Dan Heilman asked, "Isn't that like killing the goose that lays the golden eggs?" Well, since that time Microsoft has settled with the Justice Department and nine of the 18 states bringing the antitrust case. Plus it has settled with more than 100 class-action plaintiffs in a related civil case. These settlements render my bold prediction not just painfully wrong, but absurdly naÏve. So I guess some promises are made to be broken.

The gist of the online column was that Microsoft had behaved so badly in and out of court that there seemed to be no legal way for it to escape a more severe punishment than even Judge Jackson had meted out. But two things happened since the appeals court sent the case back to District for a new remedy phase. First, Sept. 11 caused a shift of focus for Attorney General John Aschcroft and the Justice Department. Second, Microsoft got the luckiest judge draw it could--District Court Judge Colleen Kollar-Kotelly demanded expediency, which ultimately won out.

In truth, I have already broken the promise once, when I praised the swiftness of the first settlement in another online column--also a colossal blunder. When I actually sat down and analyzed the whole settlement carefully, I read in horror as the result of years of legal wrangling and expense became little more than a slap on the wrist.

Actually, the settlement codifies the legality of Microsoft's predatory practices. What it takes away with one hand, it gives back with the other, and then some. For example, it does force Microsoft to share its Application Programming Interfaces (APIs) with the competition. But it also forces those who use the APIs to share their finished code with Microsoft. The result is that Microsoft would see all of its competition's trade secrets and easily replicate them. Or, though the ruling ensures that Microsoft competitors be allowed to get their icons on PC desktops, the clause only applies to companies who have sold more than a million copies of their software in the United States. The very companies who need a competitive advantage in this case can't get it.

And these two qualifiers are indicative of the whole agreement. The clincher: Microsoft would be barred from terminating a PC vendor's license agreements because the vendor cooperated with one of Microsoft's competitors. But it could still terminate the agreements, because if it did, the vendor would have to take it to court for violating the antitrust provision. How many small clone shops can afford to fight Microsoft in court? If the government ran out of money trying, not even Dell would fight Microsoft for the ability to put a Quicken icon on a Dell desktop.

Our only hope for justice is that the nine states holding out for stronger remedies move to strike several qualifying conditions that take the teeth out of the central remedies. Fortunately, two of the holdout states--California and Massachusetts--have the most to lose from a settlement that legitimizes the behavior that spawned the case. And they're no dummies. Massachusetts Attorney General Tom Reilly told the BBC that the deal was "full of loopholes and does little more than license Microsoft to crush its competition." Strong thinking like that just might put the teeth back in the deal.

If the antitrust settlement makes you cringe, consider the civil settlement. For years, Microsoft has been using its monopoly on operating systems and office software to charge about four times what a competitive-market would allow. Consider Office, which costs just less than $500 per copy for the full-featured suite. Outside of a few graphics programs, most other software costs less than $100 per copy these days. Most IT folks will agree that Microsoft gauges people who need Office by virtue of needing to share documents easily with colleagues. I'll let you do the math, but it doesn't take a Harvard economist to figure out that Microsoft and its investors have earned billions of extra dollars over the years due to its OS monopoly alone.

So what does Microsoft get for gouging consumers? It must donate a lot of old Microsoft computers and software to poor schools around the country, something it needs to do periodically anyway to keep its staff in the latest tools. All told, the in-kind donation would be worth $1.4 billion. Of course, that value is based on the assumption that the hardware and software is new and Microsoft charged its own market rates for it. Instead, it gets a chance to re-outfit its employees and look like the grand philanthropist that the image consultants have urged it to be all these years. It also gets some inroads into schools, the one market it has had some trouble in (it only owns about half that market). Then, Microsoft probably gets to write the whole thing off. And the people in the class-action suit--small and medium-sized businesses mostly--get nothing for bringing their cases to court.

The question is, what will Microsoft gain from the victories? One answer is, more contempt from the rest of the computing world. Developers the world over have long since written off the antitrust process as ineffective, and have been quietly waging a war against Microsoft. The battleground is the Internet, and the weapons are worms, buffer overruns, and other hacks into Microsoft's leaky server environments.

The situation has become more dire than one might think for Microsoft. In years past, when analysts calculated the total cost of ownership for Microsoft environments, security was a hidden cost and training was a hidden benefit. The conventional wisdom was that it was cheaper to use Microsoft overall, and these two hidden cost/benefit factors canceled each other out. Since Code Red, the conventional wisdom has changed. Analysts now say it is far too risky to sacrifice security for ease of use. Microsoft is losing the security war, and the demand for MCSEs is dropping rapidly.

It is impossible to tell how many fewer hackers would exploit Microsoft servers if the company was a good corporate citizen. But its bullying does have some effect, as the slashdot.com diatribes will attest. Unless the holdout states get a stronger settlement, the political fallout for a Microsoft settlement will be even more severe than what we saw after Microsoft's tactics were revealed in all their glory during the trial. Microsoft may have won the antitrust battle, but it is losing the security war. And there is little hope for Internet peace in the near future.