Visit Your Local PBS Station PBS Home PBS Home Programs A-Z TV Schedules Watch Video Donate Shop PBS Search PBS
I, Cringely - The Survival of the Nerdiest with Robert X. Cringely
Search I,Cringely:

The Pulpit
The Pulpit

<< [ Be Careful What You Wish For ]   |  Tactics Versus Strategy  |   [ Terminal Condition ] >>

Weekly Column

Tactics Versus Strategy: Why Microsoft and Intel Both Lost GroundThis Week to the Open Source Movement

Status: [CLOSED]
By Robert X. Cringely

Never fight a land war in Asia. Never invade Russia in the fall. Whenever possible avoid wars on two fronts. Alexander, Hitler, Napoleon and George McNamara all learned various combinations of these lessons of engagement, but Microsoft apparently hasn't, having fallen a bit further behind this week to the Open Source Liberation Army. And in a very peculiar way, Microsoft's countermeasures can't help, but make life harder for an old friend — Intel.

Some of the news was obvious, some subtle. The obvious part was Sun Microsystems' purchase of Star Office, an application suite that runs on a variety of platforms, but mainly on Linux. Star Office, while simpler than Microsoft Office has the advantages of being, well, simpler than Microsoft Office. And of course, the price is right. But even more importantly, Sun will be opening-up the Star Office source code, which should result in a thousand upgrades blooming — each one another headache for Microsoft.

To make things worse for Redmond, Sun plans a Web-based version of Star Office that will allow the cheesiest little Java Web browser to run a sophisticated office application, even without local storage. Spreadsheets on your TV set top box are only months away. And most important of all, Sun cut a deal with the gurus at Linuxcare to support both developers and users.This puts Microsoft at both a price and platform disadvantage and virtually guarantees we'll soon see server-based office applications on WebTV. The Linuxcare wildcard also adds high-quality tech support, which is the thing big business has worried about when considering free operating systems. Lots of outfits are turning to Linuxcare, which is the only Linux startup so far funded by the smartypantses at Kleiner Perkins Cufield and Byers. Buy a Dell computer with Linux installed? Dell's Linux support is really Linuxcare.

Part of Sun's strategy in buying Star is to put pressure on Microsoft and boost server sales. The rest of the strategy is to revive Sun's comatose JavaStation network computers. Have you ever known a JavaStation user? Me neither. Whatever the motivation, it's a smart move on Sun's part — especially open-sourcing the code, which will give Microsoft fits and cost Sun absolutely nothing.In a completely unrelated but equally important event, Microsoft decided to stop development work on the 64-bit version of Windows 2000 for Compaq's (formerly Digital's) mighty Alpha processor. Probably Windows 2000, which has been delayed several times already and stands to be delayed once again, was just too hard to get running on two 64-bit fronts at the same time — those fronts being Alpha and Intel's new, and equally-delayed, Merced processor. In fact, it is easy to see the Microsoft move as a love poem to Intel, though that's not at all its intention or reality.

Windows 2000 is a monstrous product that Microsoft will shortly delay again with claims that it is just better to wait until after the Y2K dust settles. Killing the Alpha version just reduces the Win2000 monstrosity level a bit — for Microsoft, not for us.So what happens to the Alpha? Compaq will still be using the chips in their Tandem Himalaya servers. Samsung is also designing Alpha motherboards that fit PC models. What's more, the AMD Athlon chip runs on the Alpha EV6 processor bus, allowing the potential reappearance of Alphas in mass-market CPU boards. But what operating system will nearly all of these Compaq and third party machines be running? Thanks to Microsoft this week, nearly all of them will run Linux.

Compaq engineers quietly admit that Linux on Alpha runs better than Compaq's own Unix. So don't be surprised if Compaq kills its True64 Unix product in favor of Linux. Compared to free, it is hard to justify the True64 R∧D investment, especially given the Titanic mindset inside Compaq these days.

Or maybe Compaq will save the best parts of True64 and put them on Linux, which is similar to the survival strategy being embraced right now by SGI. With Windows NT being dropped by Microsoft from the Alpha, this frees Compaq from any contractual arrangements with Redmond, allowing for a free push into Linux. And even if Compaq is just too overwhelmed with panic and downsizing to fully embrace Linux, every other Alpha vendor has almost no other choice.So Microsoft, choosing Merced over Alpha, has helped push away one of its biggest allies — Compaq — actually hurting Intel in the process. Not that many of us will be running either Alpha or Merced machines on our desktops anytime soon. These chips will instead be running in servers supporting all those new server-based applications like Star Office. Is a pattern beginning to emerge here?

But wait, there's more! Let's look beneath the announcements and their immediate implications to the larger picture of where open source software is headed and what it means to users. I believe Microsoft is threatened not only in the server space, but at every level of computing.

Bill Gates once told me that the way to make money in the computer business is by setting de facto standards, by which he meant proprietary standards. Microsoft has done just this, first with its languages, then with DOS, Windows, and with Office. Microsoft makes incredible profits this way, even in a market with decreasing prices. But the same thing isn't happening on the Internet, where Microsoft, for all its muscle, doesn't dominate and where there are few proprietary standards.

Microsoft missed the Internet and has been playing catch-up for years. Bill Gates can't take credit for any of the current Internet boom. Then why is it happening? This time it is because of open standards, not de facto ones, with the most important standard being the Internet protocols themselves like TCP/IP and HTML.Look at TCP/IP or HTML, and you'll see that the very reasons for their success are the opposite of Microsoft's. Both are open standards that nobody owns, that follow published specifications (Microsoft hides major parts of its product specs), that require no license fees, and that have freely-available reference implementations. THIS is what made the Internet boom possible, and Linux — as well as a number of other open source operating systems — has exactly the same characteristics.

The future of personal computing is clearly to be found in low-cost machines. You can get a free PC today by cutting an Internet service deal, but two years from now you'll be able to buy a real PC for under $100 and even the Internet service will be free. Making machines that cheap will drive them into all levels of society doing things we never expected. To appeal to the broadest range of users, the operating systems of these cheap machines will have to be hidden behind simple user interfaces yet to be designed. There is no way Microsoft can make and support a powerful operating system that runs on a $100 device. In contrast, a free operating system — ANY free operating system — can be economically deployed on a $100 platform today. That's the power of free, leaving no place for Microsoft on the user end.

Comments from the Tribe

Status: [CLOSED] read all comments (0)