15 Feb 2004

Mr Gates’ Dire Straits

Microsoft is having its problems lately, of which the news that it has managed to convince East London’s Newham Council to stick with Windows rather than shift to free software by providing unsustainably deep discounts in pricing looked like the very worst, until something even more dismal came along. Valentine’s Day found the monopoly coping with the unauthorized release of a large quantity of Windows NT and Windows 2000 source code, apparently stolen from or mistakenly made accessible by a commercial partner with a source code license. The first news item passed for a victory at Redmond, no doubt, in keeping with the “Never Lose to Linux” strategy for public-sector marketing previously propounded in Orlando Ayala’s famous secret memorandum leaked last summer. But the Pyrrhic victory at Newham Council and the Stolen Source Escapade—unrelated as they seem—represent very neatly the two blades of the scissors that are, to paraphrase Mr Gates himself, cutting the oxygen line of the tottering giant.


The triumph of poor Newham Council, which could not afford even to upgrade its existing Exchange server, let alone participate in the glorious march of technology by buying all those Windows XP licenses on the desktop, was a masterpiece of hard-nosed bargaining, and a taste of what free software competition on the desktop is going to do to the world’s most overvalued firm. By carefully cultivating an entirely credible proposal for a complete free software desktop alternative, and demonstrating an actual willingness to blaze the trail for other UK municipal governments, Newham made itself a customer the monopoly could not afford to lose.

So Microsoft “won” back its customer, and prevented the first high-visibility breach in the wall imprisoning the UK public sector. But keeping your customer by giving your goods away is not, to say the least, a marketing strategy for the long term. Unless you’re the Free Software Movement. No matter how large the fund is that Microsoft has put aside to undersell free software in government acquisitions competition, it cannot be large enough. Inevitably, questions will sooner or later be asked in the House of Commons about the Office of Governmental Commerce’s inability to get for the National Health Service the same software pricing offered to little Newham, and the jig will be up. The subsidies that governments, and therefore taxpayers, have paid to Microsoft over the years by buying poor software at monopoly prices more than account for the company’s $50 billion hoard, and if every government on earth held out for Newham’s terms, Microsoft would be history.

Meanwhile, the world has been unintentionally blessed by a chance to look at the “secret” everyone has been paying Microsoft to keep. The source code that suddenly appeared in the Net on Friday the 13th was vintage mid-2000, I am told by those who have looked, which didn’t include any of the Free World’s programmers. We wouldn’t have such a gift on any terms. Any exposure to Microsoft source code means acquaintance with what Microsoft claims is trade secret know-how, and anyone possessing such know-how—whether acquired under a restrictive license from Microsoft or by unauthorized unlicensed consultation of source—would be ineligible to work on any free software project to which such know-how would be relevant.

It was soon apparent, however, that some programmers had been reading the code: the first exploit allowing website operators to take over computers running the IE5 browser was published within 72 hours. “Security through obscurity” has a bad reputation for a reason. Free software gains security by the efforts of all the talented professional and amateur programmers who are constantly reviewing the source code. Free software is high quality for the same reason that scientific inquiry works: unrelenting peer review. But the proprietary development model means that reviewers with good intentions have to avoid the code, and they wouldn’t be able to do anything useful about the bugs even if they spotted them. When the only peers reviewing Microsoft products are those who want to exploit bugs for malign purposes, the result is that all the bugs are shallow, and the global security situation cannot but degrade still further.

Taken together, these two episodes show the causes of Microsoft’s unavoidable failure, so long as it adheres to the present unsustainable model of proprietary production. It cannot effectively compete against lower-priced goods of higher quality, and it must incur higher costs to protect its “secrets” than those secrets are worth. If it partially varies from the model, by price-cutting to compete with free, or deliberately or accidentally lowers the barbed-wire fence around its shabby “secrets,” the failure of its business model becomes apparent.

Competition is a wonderful thing, except for the monopolist. Customers and investors are becoming ever more aware that Microsoft’s business model will not last long enough to justify either current product prices or the current price-earnings ratio. The Microsoft Way isn’t history yet, but history is only a matter of time.

This column was first published in the UK in Linux User. It is also available in PostScript and PDF formats.

permalink | columns/lu | 2004.02.15-00:00.00

Comments are closed for this story.

Trackbacks are closed for this story.