Guy Kewney

It should be no secret that the European Commission is worried about Microsoft's influence on the economy and the IT community, and is investigating ways of avoiding its monopoly abuse of power.

It should also be no secret that whatever the EC actually decides is most unlikely to produce an effective way of curbing that monopoly - if it does decide it's a monopoly.

However, if I were planning strategy at Microsoft HQ in Redmond, I'd be more worried about 'market forces' and, particularly, the reluctance of corporate finance departments to pay increased software licence levies, or what I'll affectionately call the Microsoft tax, which is going to go up.

The first time Microsoft told me about its .Net strategy, it seemed just a me-too way of getting back at Sun Microsystems over the Java battle.

Sun wouldn't let Microsoft take over Java, so Microsoft decided to play with its own ball. Or so it seemed to me, but I now think I was completely wrong. The thing that Microsoft has to do - more than anything - is to grow.

It may be huge already, but its financial future depends on the confidence of investors that it will continue to expand.

Three growth options are available: increase the number of personal computers; move into new markets; and move into the IT infrastructure.

The number of personal computers, surely, is at a maximum. The number of machines sold each year may rise again, but there will never be noticeably more PCs in the world than there currently are.

Everybody who comprises this market has one, and almost nobody needs more than one. The Microsoft tax is payable per machine; therefore revenue stays the same.

There are other devices that can carry a Microsoft tax. PDAs, for example, the Pocket PC, the Smartphone, the portable Windows Media Player. The move into new markets is a gamble that is far less likely to succeed.

The phone business turns over a hundred times as many units as the PC business, but the fees paid are one per cent of the PC fee. And can Microsoft really dominate that market as it does the PC business?

The arrival of Windows Server 2003 marks the success of the move into IT infrastructure. It puts Microsoft on level terms with IBM and Sun, or it will do if it truly works.

Naturally, there will be those who argue, eloquently, that Microsoft will not be able to succeed. I think I'll save those arguments for another day, and just say that it has succeeded before, when all the pundits said it wouldn't.

But success in IT infrastructure isn't necessarily going to make Microsoft twice as big as it is today. The licence fees you get from corporate servers are traditionally huge.

But Microsoft has broken into that market primarily by charging less than the incumbents. Compare the cost of an SQL Server licence with what Larry Ellison's Oracle would ask per user, and you'll see that the earnings are set to fall on a per-seat basis in this industry.

So what it comes down to is that Microsoft can grow quite a bit on the basis of new industry, quite a lot on the basis of getting into IT, and maybe a little out of the media business; digital rights management of movies and music, for example.

Then it has to perform a miracle or, lacking a miracle, increase the tax. It has to go to the PC makers and say: 'You're paying too little per computer; pay more, pay double, pay treble. It's still a small proportion of the total cost, you know.'

Well, maybe. But corporate PC buyers don't like that. They are squirming under the prospect of having to pay the full, recurring cost of licensing Windows every year, if XP copy protection succeeds. And if the cost doubles, or trebles, the squirming will turn to squealing, and then screaming.

They will, in short, try to find an alternative. Until recently, the Linux and open source alternative was dismissed. Open source plays a small (but growing) part in corporate servers, but desktops are all Windows.

The tide is turning, however. Linux is far more secure in terms of viruses and in preventing unauthorised user-loaded software. All it really needs is Open Office to succeed.

Five years ago, I wrote that the obstacles facing Linux were huge. If you wanted to hire someone to do the accounts for a month you could easily find someone with Excel experience, but no temp agency would provide you with someone to use whatever Corel toy was current.

Is that changing? Does Open Office allow Word and Excel and Powerpoint experts to apply their expertise? Perhaps, not quite yet - but maybe soon.

If the corporate financiers squeak and squeal enough, the beach defences Microsoft is trying to set up against piracy and licence evasion could find that the incoming tide will simply go round behind them.

I think the trend for the next two years will be an increasingly irresistible pressure to switch to desktop software that carries a lower tax than Windows.

Inevitably, this will mean gradually higher fees for the open source applications, as their support costs standardise.

Maybe it will allow Microsoft to claim that it really isn't a monopoly after all, but then, I never expected that to bother Redmond much.

Like this story? Spread the news by clicking below:

 del.icio.us     Digg this     reddit!