Keeping computing's FUD factor at bay

Once, IBM salespeople were known to employ "the FUD factor" (Fear, Uncertainty, and Doubt). Salesmen would cast doubts on competing products and remind the customer: "No one was ever fired for buying IBM." In those days, standards were closed. This meant that one vendor's computers wouldn't talk to another's. Once you made a choice, you were stuck with it.

In the 1980s, open standards like TCP/IP changed everything. If Sun's Microsystem's file server ran faster than Silicon Graphics's, you'd buy Sun. If Hewlitt Packard came out with a faster one, you'd buy that, and know they all worked in the same way.

But if a vendor's software had errors, you still had to wait for the vendor to fix it, something that might take months or years. And if you wanted to add new features to the software, you couldn't, because the source (the English-language version of a program you need to make changes) was secret.

So Richard Stallman decided to create an "open source" version of the Unix operating system that anyone could modify. His Free Software Foundation never did get a full Unix out, but they developed several major programs and pioneered the concept of open source. It fell to Linus Torvalds this decade to create Linux. Working with a worldwide network of volunteers, Torvalds created a full-fledged open-source version of Unix. Now anyone can download the sources, fix bugs, or add enhancements.

At the same time, another group of developers was creating Apache, an open-source Web server (the program that your computer connects to when you call up a Web site). Soon, cash-starved Internet service providers started to realize they could run Linux and Apache on a powerful Pentium system, and have the performance, stability, and features of systems costing many times more.

Apache and Linux became the most popular Web server and Unix implementation. And, in a nice twist, IBM decided to bundle Apache as the Web server with some of its product line. Suddenly, the king of FUD was promoting open-source software.

But then a pair of leaked internal Microsoft memos surfaced on the Web over Halloween weekend. In a carefully drawn analysis of Linux, they explained why it was competing so successfully in the Web server business against Microsoft's Windows NT. They also praised the open-source movement as a highly effective way to develop quality software.

But the memos continued with a strategy for combating Linux and Apache, focusing on "decommoditizing" Internet standards. This means taking existing standards and adding features that work only with Microsoft products. Then, if developers adopt those features, the customers will need to use Microsoft products.

In short, a closed standard.

If this happened, rather than competing on a level field, Microsoft could lay the seeds of doubt in its customers. If customers didn't choose Microsoft products, they could be locked out.

FUD is alive and well and living in Redmond, Wash.

This isn't the first time Microsoft has allegedly tried this trick. Sun Microsystems is currently suing over proprietary changes it says Microsoft made to Sun's Java programming language. Netscape has similar complaints about Microsoft's Internet Explorer. Incidentally, this summer Netscape announced it was making its browser a free and open source.

So the industry finds itself at a crossroads. One path leads to a world of open standards and open sources, where software has to prove its worth. The other leads to a world of proprietary closed standards controlled by a select few. I know which one Bill Gates would prefer.

* James Turner is manager of Web-site technology at The Christian Science Monitor's Electronic Edition. You can e-mail him at james@csmonitor.com